[Unboxing/1st Impressions] HP ENVY x360 w/ AMD FX-9800P

Formula.350

[H]ard|Gawd
Joined
Sep 30, 2011
Messages
1,102
UPDATED WITH PRELIMINARY BENCHMARKS AND AUDIO IMPRESSIONS

~~~~~~~~~~~~~
I decided that this was just far too much to post as a reply in the original thread and figured I might as well give it one of its own. Much of the context of this is written with having read --this post-- from that thread, at the very least.

Not sure if one can qualify this as a review, unboxing, first impressions, or just the mostly-coherent ramblings of a tech junkie (aka user-experience). For the purpose of it all I qualify it as a First Impressions + User Experience = Micro Review. (I've never been much of a writer when I 'had' to, but I have a couple under my belt from during my time as the Editor for Pro-Clockers once upon a time. Point is, there will of course be personal bias exhibited here! That, and me being less mindful of my grammar.. lol)
~~~~~~~~~~~~~

As an update to information I posted in the other thread: Seems I misquoted specs about the hard drive on the AMD model... I could have sworn it came with the 2TB and the Intel model came with a 1TB + 256GB m.2 SSD, to the point I feel a bit crazy now that the invoice and their site both show 1TB on the AMD model :confused: :unsure:

After delays at customs, it arrived! One end of the box has definitely seen better days though, unfortunately that's the end with the laptop, not the end with only the charging brick (box is a thin rectangle vs more of the thin square that I'm used to them arriving in). So fingers crossed that no harm has came to it, as the only foam the thing incorporated was a thin (3/4in?) layer on top. Just cardboard with folds to keep it spaced apart, not even corrugated cardboard to absorb any rough ride. *grinding eye-roll*

ENVY_Box.jpg



Which upon removal from it's plastic bag and inspecting, I can't help but wonder if it was that rough ride or if it's the Grade-A craftsmanship to blame, but looks like the screen is slightly bent? With lid closed (as they come out of the box) one side has a significant gap compared to the other. Really do hope it isn't bent and opening/closing will sort it out, IOW praying it's hinge related... :nailbiting: (And yes I've taken pictures of the box prior to opening, and the gap prior to ever lifting the screen, which the fabric sheet is still in place). Pic #1 is Left side with no gap (fabric visible), Pic #2 is Right side with gap. Pic #3 is close up which shows the gap and the screen bumper, which visible above the bumper is reflection, a reflection which wouldn't be there if it were touching (has some annotation).

ENVY_Lid-Gap_L.jpg ENVY_Lid-Gap_R(way-to-go-on-that-focus-job!).jpg ENVY_Lid-Gap_R(yep-youre-fired).jpg

No cracks on the screen, so that's good. Pretty sure I figured out why most of that gap was there, and feels as though it comes down to shit craftsmanship :\ The screen (digitizer?) wasn't seated completely in the lid all along the top edge. Took pics, then gently squeezed between my fingers in multiple spots which snapped it in to place. Still a slight gap on that same side when closed, so we'll see if there's any damage after powering up. Pic #1 shows the top edge of the lid where the screen isn't snapped into the frame, while Pic #1 is just zoomed in more and gives better detail of what is what (the rubberized lip does protrude on other sides). Pic #3 is a properly (I hope) recessed edge and is roughly the same amount indicated by the bottom line in #2...

ENVY_Lid-Top.jpg ENVY_Lid-Top-zoomed.jpg ENVY_Lid-Side.jpg

The AMD FX sticker is cockeyed :( Even this shit Acer Netbook had straight stickers. Foreshame HP heh (Honestly though, it's the only sticker on the thing, which looks a helluva lot nicer than most new laptops with 10 square miles of "What's Inside!" stickers, so it does kinda bum ya out that those can be applied with enough care but this tiny sticker can't)

Onto the... um... USB Power Brick Adapter? lmao Hope this doesn't mean it'll A) Take a month to charge. B) Not provide enough power to allow the system to charge while being used in any kind of meaningful capacity. I have honestly dealt with larger SMPS wall-warts for smaller devices than this, which as you can see how the Netbook's compares!
Left to Right... Circa-2001 Dell Inspiron power brick for an early P4 era laptop. Circa-2010(?) Acer Aspire One Atom-powered Netbook. Circa-Today HP ENVY x360 power adapter...
ENVY_Power.jpg

UPDATE: For those curious, it's a 45W (19.5V 2.3A) adapter. Charging does not take forever, YAY! While plugged in and light desktop usage, it took roughly 1hr 45min to charge from ~14% battery (50% screen brightness). HOWEVER, oh my... that little thing would make for one hell of a coffee warmer! I decided to stick a thermal probe to it and ... 53.5C :wideyed: That's nearly 130F for the rest of us... Ambient was 19C (66F) BTW and I do have decent airflow in my room, so it was kind cooling the thermal probe. I figure it was closer to 55-57C if the sensor wasn't being cooled, which I can only imagine what inside temps are. All I know is that's a lot of waste heat, which if [H] and other site's PSU reviews have taught me is that heat == inefficiency and the more efficient a conversion of voltage is, the cooler devices run. So 80+ I do not anticipate lol (If anyone is smart enough to figure out a very rough efficiency % estimate of going from 120V AC to 19.5V DC with roughly 55C waste heat, I'd be interested. lol)

Last gripe: The lid... Is OPENING a laptop's screen an afterthought these days? What exactly do they think about when designing this aspect? lol Sure they provided a little lip for you to get some purchase on with your finger but then what? The thing is light and all you end up doing is lifting the whole thing up! There's no edge around the bottom half to grab at all, it's contoured into a thin meeting point so unless you have a bit of fingernail to hook onto, you'd need a some sort of pry-bar heh I seriously may need to just go "Form over Function" and take some packaging tape to stick on the keyboard palm rest with a folded end just so I have something to hold onto :rolleyes:

.....

And power up... Err... Ok it actually took me 2minutes to find the power button :oops: I feel like I'm using a giant cellphone, cuz it's on the side next to the Kensington Lock hole.
Anyways now we're powering up, and *whistles* ... that's some lovely light bleed along the top of the screen (naturally only noticeable on a black screen like during boot up). I sure do recommend to anyone setting up Windows 10 for the first time to not use the "Express Setup" and do Customize. That appears to be where a LOT of that tracking crap is enabled at :yuck: Of the 12ish options to turn something On/Off, the only one I left "On" was the one to Send Error Reports to Microsoft! lol The rest were all of that "Send usage of this and that", "Make browsing better by pre-loading things we think you'll view", "Connect to this or that sort of hotspot, or because your contacts connect to it", etc etc.

In general, it works well so far. Something to keep in mind is that I'm old school, and am kinda set in my ways on certain things, so some of my opinions on this may be overly-critical. For example I love the keyboard on this Toshiba I'm replacing (and typing this on), as it has the rubber-nipple with scissor-lift keys. Granted, sometimes crap ends up getting in the scissor-lift mechanism which binds up and causes depressing to be tough (usually it's a dog hair, yay Huskies!) but it is very tactile and... well the keys you can tell you're on them because they have a bit of wobble, compared to this Chiclet keyboard on the HP and likely every other modern keyboard on laptops (maybe save for gaming models). Next is that I use a touchpad for it's most basic functions: move the pointing, right-click or left-click. I once again prefer the tactile physical button style touch pads, and not so much these multi-point touch modern variants.
For example: I had bought my mom a new Dell last year for Mother's Day, as she wanted something with a big enough screen and oomph to handle Photoshopping her DSLR pics. I picked up a 15" and had gave it a trial run while setting it up so that she could just turn it on w/o the setup. I tell you... that touchpad is unusable out of the box, and barely so after I futzed with the settings (Windows 8.1, if it matters). First off it would screw up with Palm Detection turned ON, so that had to go. Secondly the lack of physical buttons means clicking while selecting just is a pain due to touch detection. I don't know if it's the drivers, Synaptics software, or just the hardware, but I have SO many issues trying to click and drag to select things or drag and drop thing. If I'm trying to CTRL+Select multiple items it usually ends up unselecting everything when I click to drag said selected things, only dragging the one item. Also I've noticed a lot of times that when I had been typing, due to the style I developed, a lot of times the S or A won't register due to the Chiclet design, which I've never experienced with this Toshiba.


TOUCHPAD & KEYBOARD: That being said, I'll start by saying that the touchpad on this ENVY is like most modern laptops in it's multi-point, lacks L/R physical buttons and has a single center-click button behind the whole pad. There is a general "cheap" feeling to it when you tap-click on things, due to the fact it has that single button behind the entire touchpad. To be clear, this is a tap-click and not a depress of that button, so the cheap feel is due to a sort of hollow-sound combined with a little clunk as a result of my finger tapping the touchpad on account of the pad being physically free-moving (on 3 sides I think). Compared to this Toshiba which the actual touch section is built into the palm rest, tap-clicks have a understandably solid feel to them. Again this could just be that I'm being overly-critical and isn't really what I perceive it to be, so in the end I'll just have to get used to it. Typing on it seems OK so far, albeit the limited amount I've done on it, none of which has been typing anything beyond a few words for input boxes.

Pre-Posting update: Been using it a bit more and so far the keyboard is kinda hampering me. It's going to take quite a bit getting used to. I might also have to see if the Intel model's backlit keyboard (I think someone mentioned it has one?) will work on the AMD model as that'd help quite a lot in terms of getting my hands accustomed to landing on the right place.
I'll be damned, it DOES have a backlit keyboard! Fn+F5 turns it on lol As for the touchpad, it's not anywhere near as bad as the Dell's, but I am rather annoyed with it's tap response :\ I'd say roughly 25-30% of my taps aren't registered as a click due to being a bit too soft, which is plenty enough of a tap for this Toshiba. Honestly I blame the fact that there's SO much software driving these newer touchpads that it comes down to interpretation of what is a click, what is an accidental brush, what's a palm, multi-touch response, etc etc which causes it to be less of a mousing surface and more of a bastardized touchscreen.


SCREEN:
Is a 15.6" IPS 1080p Touch. The screen is decently bright on the max setting (though definitely not as bright as this Acer One netbook, holy crap is that thing blinding), and dims down fairly well at 0%. I feel 50% is a comfortable indoor-with-lights-on level. I'd say it's a semi-gloss but has no anti-reflection coating on it. View angles Up and Down, the brightness is retained well and doesn't appear to exhibit any quality degradation. Viewing Side to Side there is noticeable dimming but not too significant, retaining readability pretty much until you're looking right across the panel (I suppose that's what they consider 176deg?), in other words if you and someone else were not viewing directly on while watching a video I think both parties would be equally happy. Perhaps worth noting... While messing about in the settings for Resolution I noticed the Radeon control panel display a dialogue window on the corner, which first one I think mentioned the ability to connect to another screen wirelessly if you wanted, which is nice but not the "worth noting" point. The second popup I didn't get time to read fully but... I swear it said "something-something FreeSync display detected(enabled?) something-something display port(?) something-something" o_O Checked out the Radeon Settings, looked at the notifications, and sure enough... "You have connected a AMD FreeSync™ technology (or DP Adaptive-Sync) supported monitor. Use Radeon Settings to enable FreeSync™" so WOOT! I mean sure, not a huge deal given this isn't going to be a gaming powerhouse (*grumbles* thanks to no dual-channel RAM), but to me it's a fairly decent selling point that they should've mentioned on the any number of the pages! When hovering over the FreeSync option it seems to have a range of "40 to 60 Hz", which is good enough for a laptop IMO.


PERFORMANCE: To be conducted Slight concern though, as I noticed when in the Radeon Settings that the memory was clocked at "933MHz", which equates to only DDR4-1866 and not the advertised DDR4-2133! May be a 2133 DIMM, but it doesn't seem to be clocked at that. I'll be sure to contact HP and request a BIOS update if it turns out to not be running at speed. Might need to request a secret DRAM menu to allow us to select the frequencies supported by Carrizo, given I had considered getting a 16GB stick of DDR4-2400.

First is Geekbench3 (BIOS F.09, in case I get them to update): HP HP ENVY x360 Convertible - Geekbench Browser

First up Next up is AIDA64. However, since I've reached the 10 image max on this post, I'll have to utilize the second one :) Here's the CPUID image though, which shows that it IS unfortunately running at DDR4-1866 :unsure: Shame on you x2 HP!
cpuid ENVY x360 FX-9800P.png


BATTERY LIFE: Still running these tests... It did come with a 70% charge, and during my setup had jumped around between 3h 45m remaining and 7hr Xmin remaining. Screen brightness had defaulted to roughly 80% and I assume my having knocked it down to 50% will help a bit more. I also created a Balanced profile (to replace the "HP Recommended" labeled profile), otherwise having all the same settings as the HP but with a Passive fan setting on Battery (tweaked the Low/Critical % levels as well, seemed a bit higher than needed). As for anyone curious, it's a "4 Cell" with a capacity (AIDA reported) of 55,671 mWh, and I presume that it also will have no aftermarket extra-capacity batteries available, given the internal battery design.

Pre-Posting Update: Been using it for a few hours now. Still just general putzing about on the desktop, no internet or other stuff. It's down to 24% and reports roughly 2hr usage left. Figured I'd run it out before I connected it up to that beast of a charger... lol

Update: Watched Robin Williams - Live on Broadway on YouTube, which is 1h 40m long, albeit only in 360p quality. Decided I'd continue using battery instead of plugging it into AC, which when I started off there was roughly 35% remaining charge, giving me roughly a.... 1h 50m to 2h remaining time! (It was roughly 3.5hr before started watching) I also decided to keep the battery box open to keep and eye on how this all was effecting things. Again, this is with a "Balanced" profile which for Battery had Screen @ 25% Brightness, Min/Max CPU of 5%/100%, Passive Cooling, and basic other "Balanced" configured options. Worth noting was that for about 1/4 of the time I had the keyboard backlighting on. When it had trickle down to about 25% with 1h 16min batt left I decided I'd turn it off and while the % didn't change any, I was surprised to see the time rise to having 1h 24min left. At any rate, Windows only lets you set the "Critical Action" % as low as 7, which I wanted to run the thing out completely, since at 8% left it was reporting 34minutes of usage left by the time the show was over.
Whether or not one can consider this good or bad, I don't know really. However, I do plan on seeing how it does when watching something with a bit higher resolution.

AUDIO QUALITY: Preliminary results... Gave some music a listen last night and there is good news and bad news...
The BAD news is, either the laptop's speaker setup is lightly damaged, or "Bang & Olufsen" screwed the pooch on designing the audio for this thing. Usually when you listen to audio from something, unless it's mixed to have channel offset, you expect the audio to be received basically equal to each ear. On the ENVY that is not the case... The audio was extremely Right-heavy, and in order to actually make it sound somewhat balanced I had to change the balance to 66R | 44L! Even then, it could've probably gone down to 42 on the Left. There was a noticeable difference in what was being produced through the L channel compared to the R.
The GOOD news is, the audio that was delivered was very good! It was rich, and warm, neither are things I expected. I believe it has a "2.1" configuration, which the amount of bass produced would make me feel that there is indeed a lil Sub hiding out in there. Furthermore, listening through headphones (earbuds), I was quite astonished by the sound quality it delivered. Basically, it comes across as the same rich and warm audio as through the speakers. Also, confirmed that it isn't subsystem related as both channels came across perfectly normal.

I listened to How To Destroy Angel's "An Omen" 6-track EP (320Kbps MP3s) , which is sort of a mix between Pop and Industrial (fitting, given it IS Trent Reznor's wife singing, with Reznor, Atticus Ross, & Rob Sheridan). After that I gave The Chemical Brothers album "Dig Your Own Hole" a go (FLAC), which for anyone who doesn't know them you're living under a rock and are missing out, so go utilize YouTube!
Since I don't yet have Foobar2000 installed, I just used MediaPlayerClassic in an effort to have no sort of audio enhancements being used beyond whatever Bang & Olufsen have messed with through drivers. However, do note, that their audio control panel disables any of the 3 presets you can select when using headphones: Music, Voice, Movie, and none of which fixed the aforementioned speaker issue when disabled.

Overall I'm really happy with the audio output from this thing, so much so that I may be able to retire my old Galaxy S1 that I purposed as just a media player, as I loved the sound it delivered with VoodooSound enabled Kernel thanks to the Wolfson DAC. There was PLENTY of volume headroom while using earbuds, as I had it at 24 which I considered comfortable. I'll give the Sennheisers that came with the Xonar a go in a bit to see how they are handled, but I'm sure it'll be fine.


Anything else I've missed, or anyone is interested in, I'll add later. Keep in mind that running tests which require big downloads (like 3DMark) I'm going to have to decline on as that'll just eat up too much of my precious 18GB/mo :\ (plus, my internet is slow as hell most times, sometimes as fun as 20-30k/s)

Enjoy :)
 
Last edited:
AIDA64 results...
Bit of a let down to be honest. Had hoped for a bit more pep, especially from the memory, but maybe a little more performance can be had if I can get HP to correct the memory speed to actually run @ 2133. I can also provide screenshots of each run's result window in order to provide a better comparison than what I have, which is this old Toshiba (and to be fair, it hasn't been restarted in almost 2months, not that I think that'd have helped the results too much).
First is Comparison of the two. Second is Mem/Cache. Third if GPGPU.
ENVY_AIDA64.jpg cachemem ENVY x360 FX-9800P.png gpgpu ENVY x360 FX-9800P.png

EDIT: Just as a bit more context... I run an old A8-3850 on my desktop, not because I don't have anything better, just because... well I love that chip for some reason lol Anyways the Llano has, IMO, AMD's best memory controller. I'm running a silly combo of a high-speed 8GB and 4GB kit together, @ DDR3-2000 9-10-10-28 2T (they're a 2400 and 2200 kit, respectively). I have the base @ 107MHz which results in a NB of 963Mhz and CPU of 3.1GHz.
Its performance is... Read: 22GB/s, Write: 22GB/s, Copy: 18GB/s or 20GB/s (varies run to run). I know my A10-7850 didn't come close (I didn't have much time with it before a BIOS flash bricked the board), but I guess I was hoping that even single channel this FX-9800P would get 12GB/s across the board.

CPU-Z's Bench (more as a novelty)
CPUz_Bench-vs-TheWorld.png

While not exactly the most in-depth benchmark, nor does it indicate what exactly the test is or the numbers mean (beyond being an easy means for comparison), I was a bit surprised by the turn out in comparison to the Intel chip. Granted that's an Ultra-Low Voltage model, it's still clocked at 2.8GHz. However, full disclosure: I had set the 9800P's AC Profile with 100% CPU on Min/Max, so it was running at its full boost clock rate of 3.6GHz.
 
Last edited:
I am disappointed that there is no picture of the crooked sticker.
 
Please forgive me! I'll be sure to get one :D lol (srsly, I'll try to get it after Geekbench is done heh)

Thanks, it'll keep me up at night if I don't see just how crooked it actually was.

Joking aside, I'm looking for a laptop in this price range for some lower end mobile gaming, so I checked this thing out but it seems like a used last-gen laptop is a better deal considering Carrizo CPU performance and Bristol Ridge isn't really an upgrade.
 
Thank you for sharing!

Could you run Geekbench [1]?
You're welcome :)
Here be the results: HP HP ENVY x360 Convertible - Geekbench Browser

Thanks, it'll keep me up at night if I don't see just how crooked it actually was.

Joking aside, I'm looking for a laptop in this price range for some lower end mobile gaming, so I checked this thing out but it seems like a used last-gen laptop is a better deal considering Carrizo CPU performance and Bristol Ridge isn't really an upgrade.

ENVY_Sticker.jpg
Honestly, it's not that bad, but sure does look worse to the eyes. Either way for $700, I expect perfection! lol

As far as things go, I am slightly disappointed in the performance as far as what benchmarks are telling me, but I'm hoping BIOS updates can correct some of it, so we'll see.
I mean, I had some hopes that the DDR4 was going to kinda make Carrizo shine even with mind-numbingly stupid decision to make it single-channel on this thing, so far it feels like there'd have been no change in performance between it and DDR3. Nevertheless, we shall see how it fares against the Dell once I get Steam on here and transfer over GRiD 2, since it has a built in benchmark. Specs on it are I think an A10-7800, again single-channel, but otherwise fairly close GPU.


I'll be updating my main post with AIDA64 results here in a moment.
 
mind-numbingly stupid decision to make it single-channel

This is such an interesting trend. AMD has been hit by this pretty had due to the need for memory bandwidth by the APU, but this is getting much more common.

Take the Dell XPS 15. Over $1k laptop, Skylake i5 w/ 960M. Ships w/ single channel until you move to 16GB. o_O

We must have crossed the point where an 8GB stick costs less then 2x4GB sticks. Maybe it's only cents but all the manufacturer sees is dollar signs: switch to 1 stick, save some pennies, total components reduced by 1, reliability slightly goes up, multiple by 100k of units = exec bonus.

I mean it's nice for upgrading, so with an 8GB DDR4 stick being just around $30 - I think that would be a day 1 purchase. What's the model of the current memory stick?

Any chance of doing a CPU-Z validation?

EDIT - Didn't realize it was single slot :(
Manual has some interesting info: http://h10032.www1.hp.com/ctg/Manual/c05085451

One SODIMM slot, non-accessible/non-upgradeable
Support for DDR4 1866 dual channel
Support for DDR4 2133 single channel

Support for up to 2.0-GB maximum on-board system memory: 2048-MB (256-MB × 16 × 4 pieces)

Support for up to 10.0-GB maximum system memory in the following configurations:
10240-MB: 8192-MB × 1 + on-board 2048-MB (256-MB × 16 × 4 pieces)
8192-MB: 8192-MB × 1
6144-MB: 4096-MB × 1 + on-board 2048-MB (256-MB × 16 × 4 pieces)

So it has a single hard to get to SODIMM slot with either 4GB/8GB, and it might have 2GB soldered on the board? Dual channel mean 1866 speeds, single channel goes to 2133?

Double EDIT - And the M.2 slot can be used "only on computer models equipped with an AMD A9-9410 processor" What? :eek: What did HP do? Only thing I can think of is the FX-8300P has a larger cooling solution which blocks the M.2 location???

That manual also has "FSB" numbers that look like they're actually memory speed numbers.
 
Last edited:
Maaaaaan....how lame. I was really looking forward to this laptop but...Single channel? DDR4 1866....Freaking HP going and messing up stuff. Btw, i was looking through the manual. Did you notice that in the exploded view towards the end of manual that not only do they show TWO SODIMMs but almost the entire labeled description chart is wrong? Everything is miss-labeled.
 
This is such an interesting trend. AMD has been hit by this pretty had due to the need for memory bandwidth by the APU, but this is getting much more common.

Take the Dell XPS 15. Over $1k laptop, Skylake i5 w/ 960M. Ships w/ single channel until you move to 16GB. o_O

We must have crossed the point where an 8GB stick costs less then 2x4GB sticks. Maybe it's only cents but all the manufacturer sees is dollar signs: switch to 1 stick, save some pennies, total components reduced by 1, reliability slightly goes up, multiple by 100k of units = exec bonus.

I mean it's nice for upgrading, so with an 8GB DDR4 stick being just around $30 - I think that would be a day 1 purchase. What's the model of the current memory stick?

Any chance of doing a CPU-Z validation?

EDIT - Didn't realize it was single slot :(
Manual has some interesting info: http://h10032.www1.hp.com/ctg/Manual/c05085451


So it has a single hard to get to SODIMM slot with either 4GB/8GB, and it might have 2GB soldered on the board? Dual channel mean 1866 speeds, single channel goes to 2133?

I may or may not have mentioned it in this thread, but that's correct it only has ONE DIMM slot :depressed: What it states by that "Dual Channel @ 1886" it makes me curious what on earth it means, and makes me feel like that manual is a shared one between the Intel and AMD models. The INTEL model has Dual Channel, it comes with 16GB.

The biggest trend I see is Intel models get pampered while AMD models get random parts thrown together, disregarding anything AMD may suggest for optimal performance. I mean seriously, it feels like they make the AMD models by grab-bag. 1080p models that have been available that aren't high-end MSI or ASUS? A couple with low-performance Jaguar chips. Why? Who knows. High performance model APUs with only one memory slot? That's a trick question... no laptop these days with any respectable processor should have single channel! (and no, haters, some AMD chips are respectable lol)
 
Maaaaaan....how lame. I was really looking forward to this laptop but...Single channel? DDR4 1866....Freaking HP going and messing up stuff. Btw, i was looking through the manual. Did you notice that in the exploded view towards the end of manual that not only do they show TWO SODIMMs but almost the entire labeled description chart is wrong? Everything is miss-labeled.

If it shows 2 DIMMs, and with what Track Drew said... maybe I'll just have to yank the back cover off and confirm for myself... I'll try and do that right now if I have the right sized Torx driver (I think I do).

EDIT: :facepalm: So yea, suspicions confirmed... But first let me get out of the way, (y) to HP on making it a pain to get into, and for hiding 2 screws under the rubber pads. That being said...
A huge (n) to HP for being like Apple and being stupid. It does indeed only have ONE DIMM slot (I mean, it's 99% certain, I didn't take the board out but there's also not room on the other side). It does indeed have provisioning for on-board RAM. There was plenty of space to put in another SODIMM slot, but whatever.
Another (n) to HP for being dbags by having the provisions for an mSATA slot on the motherboard, but not even spending the extra 2.7Cents to populate it so that we COULD add an SSD.
A final joint-(n) to HP and Bang & Olufsen for being utter tools by offsetting the speakers to one side, thereby destroying the audio balance in Laptop-mode. It does indeed have 2 speakers, it's just one of them is in the damn center, while the other is to the right side in the usual spot. I mean, it'll sound fine in Tablet or A-frame mode because the speakers are now down-firing or faced away, thus the audio will not have a direct path to your ears, so I'll just have to get used to it.


Dear AMD,
Please do your loyal customers a solid by releasing your own laptops and tablets, since at least you know how to properly configure them to better utilize the processor you've created by not watering things down!
Thanks,
Your Beloved Fans!


Honestly, I'd actually rather have had it been 8GB of ALL onboard RAM but configured in Dual-Channel, as apposed to this single stick nonsense. It'd be easier for me to live with not being able to upgrade the capacity, versus having little reason to bother upgrading it. Similarly, there was no excuse to not populate the mSATA, even if we weren't given the choice of having one preinstalled. Nevertheless, light-gaming performance withstanding, it's kinda looking more like a $600 product than a $700.

Overall, do I like this thing? Yes.
Do I regret buying it? No.
Am I really annoyed at some choices made? Goddamn right! :shifty: lol


Tiny EDIT: Just had a thought as to why they designed this board the way they did, with the option to have on-board DDR4 and mSATA... To me that really sounds like the groundwork for a Tablet-only model. The SODIMM on this is by far the tallest thing, since the CPU isn't socketed. Next would probably be the 2.5" HDD, leaving the battery and CPU's Heatsink. mSATA slot takes care of the HDD and the on-board takes care of the SODIMM, leaving a device with half the thickness.
 
Last edited:
Sent in an HP Support claim about the memory speeds last night:
BIOS is not properly configured to run RAM at the correct (advertised) speeds. Advertised speeds were DDR4-2133 (1066MHz), but system is operating at DDR4-1866 (933MHz), which results in a fairly large performance impact, particularly for graphics.
Current Configuration: DDR4-1866 @ 13-13-13-31 44clk
Memory SPD Profile: DDR4-2133 @ 15-15-15-36 50clk

I request that the memory speed be corrected & either a hidden menu (ctrl+F1) for memory adjustment OR auto-selection of highest performance JEDEC SPD profile.

I used every last character available in the 512-limit on the initial claim, which I must admit, their online form was very problematic. Kept telling me the Serial Number wasn't right, and then that the Product Number wasn't right. Three browsers and 30minutes later I finally managed to get it submitted lol (which in the end still apparently didn't go through correctly as it reported I had 2 different different laptops from this product line *shrug*)

There were only 2 choices after that: Chat with a representative, or receive a phone call... no email route anymore I guess. Given it was midnight, a phone call wouldn't have gone over well, thus text was the go-to way to ensure everyone is understood correctly. Basically I started off letting the guy know I didn't really need any support, just that my message needed to be forwarded to the BIOS Engineers and they'll understand what I've put (he initially had been confused as to my problem, which is understandable). He acknowledged that I was requesting it be forwarded to another department, requested the best time that they call and I let him know. He thanked me for being "so nice to [him]" and that was that! :)

Today the Case Manager called to verify what I was asking, and so forth, which I did. Ultimately it boils down to this: They want me to send this in for service, no delivery charge on my part. I told him that that wasn't really necessary given the fact it simply needed a BIOS update since I was quite certain it was affecting every AMD model, so I'd be more than happy to perform (and lets be honest they package them to do it all painlessly). Unfortunately he said that due to this all being covered under warranty stuff (actual word was "insurance"), it'd need to be sent in so that in case anything went wrong it'd be covered and serviceable right then and there. When asked for a turn-around he wasn't really able to give one, but I pushed since I leave for the cabin July 6 for a month, and he didn't think it'd come back in time (~3weeks) :cry:

In the end we decided that I'll get back to him when I return, which he sent me his direct-extension number via email, and we'll take it from there.

---

I'll update the main post with this as well:
A bit more time with the keyboard on that and I'm beginning to realize some of the quirks it has. In short, I'm really not that pleased with the performance of these chiclet-style (aka island style) keyboards, as they are very finicky in responding to your presses, requiring you to press pretty close to directly on top of them before they register...
I made a LOT of typos on that keyboard, partially a result of still getting used to the thing, but partially due to mechanics behind the design. For example, I noticed when I would press on the edge of a key that (the A key, whether or not it matters), even though it WAS depressed and hitting the bottom, it wouldn't register. I was able to lift, press, lift, press about 4 times and every time it was the same. If I rocked my finger just a hair to the left, it would register. I'm hoping that once the keys begin to wear-in, that they'll be a lot more forgiving; thus, I'll not be so frustrated with it due to all the typos that are made.
Additionally, and whether this is due to chiclet design, software, or hardware poll rate, well I don't know. Nevertheless, I've noticed when typing on that thing there is the frequent problem of "shift" being applied to the second letter in words. It results in constant corrections due to sentences starting out like "ACtually", "WHen", etc.
The final hiccup I've been noticing is lots of double-registers of "S" being typed, sometimes resulting in words having three when they only should have two.

TL;DR - Quirks with Chiclet-style keyboards (not so much HP being at fault): the keys are very anal about where you depress them, as it may not register what you've typed otherwise; some sort of issue with shift which causes additional letters to receive capitalization; seems the 'S' key registers an additional 'phantom' press sometimes which results in an extra S being typed.

---

Lastly, just as an aside, I had grabbed the latest CPUz to check to make sure it had been reporting DDR4-1866 as well and not some oddity with AIDA64, which is when I noticed... they now have a little Benchmark thing built in! lol I decided to run it and as such have those results I'll throw into the 2nd post. There's 3, but they're just the same results again different comparisons processors. (only has like 9 comparison chips, so I picked the most relevant)
 
No, I haven't ran MemTest on it. I assume you're suggesting it to have it report the memory speed? Based on the memory performance, I'm sure that CPUz and AIDA64 are both reporting the correct memory speed. If it was something really obscure like 983MHz, or 1005MHz, I might suspect something isn't reading speeds correctly. 933MHz is bang on 1866, so it'd be a bit curious for it to be reading it wrong IMO.


I wonder if someone at HP read my thread and agreed that the price was too high... heh
ENVY_Ugh.jpg

Now I just need to figure out who I need to talk to in order to get $100 refunded to me...
 
No, I haven't ran MemTest on it. I assume you're suggesting it to have it report the memory speed? Based on the memory performance, I'm sure that CPUz and AIDA64 are both reporting the correct memory speed. If it was something really obscure like 983MHz, or 1005MHz, I might suspect something isn't reading speeds correctly. 933MHz is bang on 1866, so it'd be a bit curious for it to be reading it wrong IMO.
I wonder if someone at HP read my thread and agreed that the price was too high... heh
View attachment 4537
Now I just need to figure out who I need to talk to in order to get $100 refunded to me...

hehehe

Memtest might spit out something useful and is pretty universal :)
 
*shrug* I'll give it a go. :p


As for the price change, there may be hope yet. I was painlessly reimbursed the $100, so maybe the BIOS update will be equally as rewarding!

On that note, at $579 I feel much more comfortable recommending it to some of you who were reluctant. Granted, I still have yet to perform any real gaming (I did play the original NFS Most Wanted last night, graphically maxed at 1920x, but it's so old I doubt anyone is interested lol Played fine though), but it's nice otherwise in the other regards :D I mean, I can't really hold the keyboard against it since that's going to be the style regardless of who you buy.
 
Yea, I really really do need to try and get some games on this thing that are of relevance lol Only reason I've yet to is due to how I have to go about it... As I mentioned I don't have capable internet, and so I can't just easily download them over steam. Similarly, making a backup of them is also a bit slow-going I"ve found, so I make registry backups, and then copy the folders over the network to the other computer. Import the registry, and then (y) I just need to get my ass in gear and do that is all!
 
But are benchmarks not screwed up because of the ram speed when using the APU ?
 
I've decided it's not worth bothering using MemTest to check speed. Besides what AIDA64 and CPUz reported (933MHz), I also noticed that Task Manager and HP's own Support software corroborate the 933MHz speed. That's 3 independent sources that all say the same thing (I'm assuming AIDA's readings are using CPU-ID's CPUz code)

But are benchmarks not screwed up because of the ram speed when using the APU ?
Can you explain what you mean by "screwed up"?

If you mean that the results won't be quite accurate, then you'd be correct, since the memory performance is suffering right now. There's the very real possibility that they won't "fix" this, and it'll provide valid results. Similarly, it'll still provide us with the ability to know whether or not the memory speed is a factor in the performance shown if they do fix it, and if they add in further support like I'm requesting (for 2400MHz, which Carrizo is capable of) then it'll show us whether or not upgrading the RAM would provide any benefit. I mean if it'll end up supporting 2400MHz, then I'll gladly grab a 16GB stick of the stuff! :p
 
The higher the ram speed for the APU the better graphic results you get from the graphics part since it has no dedicated memory ....

Memtest will run without windows ;) that is the main feature of it ;)
 
Gotcha, we were indeed on the same page in terms of what you meant with the performance. Indeed, the graphics performance will not be as high as it should be, due to 1866, but it'll still provide relevant info for us. :)

I wish there were more BIOS editing tools leaked :( This thing runs InsydeH2O (with the old Phoenix interface which sucks), but the EZH2O2 editor doesn't seem to want to open it :\ Then again, the files in the HP exe are two 8MB BIN files, similarly named. So I don't know if those are the raw files, or inside some sort of container.

FINALLY at long last I can present you with a game comparison!
System #1:
Dell Inspiron 15 1366x768
A10-7300 @ 1.9GHz (3.2 Boost)
8GB of DDR3-1600 (Single Channel, single slot)
iGPU: Radeon R6 (Spectre) ROPs 8, TMUs 24, Shaders 384. Pix Fillrate 4.3GP/s, Texture Fillrate 12.8GT/s
GPU@533MHz, 1024GB RAM@800MHz 64bit Bus for 12.8GB/s (theoretical)
Available RAM to System: 7GB

GRiD2 results:
Total Frames: 4,305
AVG FPS: 32.27
MIN FPS: 26.24
MAX FPS: 43.29

I played through quite a bit of GRiD2 on this system last year, and I spent a lot of time getting it configured to actually be playable, but not look ugly as sin. The settings are all Custom but I'd give it probably a rating of slightly above Medium. Some are a little lower, but then others are a little higher. As you can see I was aiming for about 30fps and that was achieved. As such, I was very very curious to see what the ENVY could pull off! (I also ran this system through AIDA and memory performance was indeed lower, so was overall CPU performance, even though it was benched at the boosted 3.2GHz)


System #2:
HP ENVY x360 1920x1080 FreeSync (40-60Hz)
FX-9800P @ 2.7GHz (3.6 Boost)
8GB of DDR4-2133 @ 1866 (Single Channel, single slot)
iGPU: Radeon R7 (Wani) ROPs 8, TMUs 32, Shaders 512. Pix Fillrate 6.0GP/s, Texture Fillrate 24.3GT/s
GPU@758MHz, 512MB RAM@933MHz 64bit Bus for 14.9GB/s (theoretical)
Available RAM to System: 7.5GB

GRiD2 Results #1 (dirty, system being used, lots of documents open in background):
Total Frames: 5,483
AVG FPS: 41.16
MIN FPS: 32.70
MAX FPS: 51.12

Results #2 (freshly restarted):
Total Frames: 5,523
AVG FPS: 42.97
MIN FPS: 33.50
MAX FPS: 53.28

All graphics options were identical to the first system. I literally cloned the game and configs to the ENVY. The FreeSync does indeed become obvious! It is definitely apparent when it comes in and out of the range, but thankfully when entering below 40 it doesn't really tear much and so there's not much to notice. Where I could really tell was on the loading screen (it has a lot of animation and changing background) where the frame rate varies from as low as 12 to over 150. It's in the 80s-90s that you can really see the tearing though, which on a different occasion I had had turned on the Frame Rate Limiter set to 50 which kept it in the sweet spot and the loading screen was so fluid... it was a strange experience for sure lol

As you can see though, going from these two systems there is quite a performance difference. Granted, that R6 iGPU is not as powerful as that in the R7 in the ENVY, and the CPU's non-boost clock is quite a lot lower, it's worth pointing out that the Dell is configured with the GPU set to use 1GB of RAM compared to the 512MB on the ENVY. How much of a difference that makes... well I wish I could tell you, but neither system provide you with the ability to adjust it.


On that note... While trying to figure out how to calculate the fillrates again (GPUz on the ENVY didn't factor in the full clock speeds), I came across a curious and perhaps rather unfortunate bit of info regarding this FX-9800P :cry:
AMD-Bristol-Ridge-FP4-Family.png


Note the 9800P's max DDR3/4 speed :\ Seems perhaps the IMC is locked at 1833? <_> I mean I sure hope not, and it'd really suck if so, but I don't really know how AMD works the mobile parts and whether or not the controller is locked or not. I really really wish I was able to be put directly in touch with their BIOS department, instead of being put in contact with the CS rep. I might try to request from the guy if he could do that for me, so I can try and find out if indeed it's worth sending in the laptop when I get back from vacation. I'd really rather not have to if it's just going to read a BIOS dude's office, and get kicked out because "Nothing we can do!"... :unsure: Being limited to 1866 sure doesn't make me feel like it's an "FX", but fingers crossed it isn't a case of things being locked

Nevertheless, at the end of the day, the performance is a bit more than I expected it to be compared to that A10-7300, so that's cool at least. :D

[Edit: Just fixed an error on the ENVY's R7, I had left it as (Spectre) but its codename is Wani.]
 
Last edited:
Oh not the old AMD laptop with single channel ram again in 2016?

I was always a bit disappointed with my 2005 Compaq laptop that had a full size Athlon 4000+ Venice CPU in it, only had single channel ram.

Was still a bit of a beast at the time though.
 
Oh not the old AMD laptop with single channel ram again in 2016?

I was always a bit disappointed with my 2005 Compaq laptop that had a full size Athlon 4000+ Venice CPU in it, only had single channel ram.

Was still a bit of a beast at the time though.
*sigh* Yea. As I mentioned [tinfoil hat mode]:bag: I really do feel as though it's still pressure by Intel to OEMs to do that. It happened before, I don't see why it couldn't be a persisting trend. While Intel's iGPU has increased in its capabilities, I don't really know if it is a match for a Radeon (except for maybe those Intel models with eSRAM for the graphics). That's the biggest reason I could see as to why there would be a desire to cripple the AMD system with a single stick. *shrug*[/tinfoil hat mode]

Nevertheless, from what I've been seeing, Dual-Channel mode is 'enabled', just not active. HWiNFO reports that Channel B is tied to the on-board DDR pad (unpopulated of course), which unfortunately doesn't really do us any good when they don't offer the build-option to GET it populated. :wtf:

Alas, it's far easier to complain than it is to praise, especially when said complaints are about something that are so silly that they shouldn't have been that way in the first place :LOL:
I still do not regret this purchase! :happy: Maybe when the warranty runs out I can find someone with the know-how, SMT hot-air tools, and willingness to solder some DRAM chips onto this board for me (and maybe the NVMe/m.2/mSATA/whatever slot, assuming the other hardware bits are there and just not the actual adapter) :cool:
 
I've just now gotten around to actual gaming, not just tests, and I may have some woeful news :confused: Been playing GRiD2 and I don't know if it's Windows 10 related, or what exactly, all I know is twice it has force-closed with no explanation at all. One moment it's chugging along, seemingly alright, and the next thing I know I'm back on the desktop. No error, just as if you ALT+F4ed mid-game.

I had left GPUz open this second time to see if it's thermals related and... well I dunno if it's accurate to be honest lol For the Sensors it can't read the Core clock at all, Mem clock is "56Mhz" can't read the Load either. Not sure if Dedicated RAM is being read right but I think so given it never exceeded 512MB (439 was max), but the max reported temp was 72C but around 66C when things suddenly quit.

I could've sworn I had seen that HWiNFO had an experimental fan speed option, but I can't find it for the life of me... I was going to see if it worked so I can crank the fan to max then re-test. Oh well :\

Alternative theory that just came to me: Power Adapter related perhaps? Not entirely sure how though.
 
No, I have not updated the drivers. They are Crimson 16.1.1, driver package 15.301.1811-160120a-300958C-HP (which seems to indicate a build time of Jan 20th, so fairly old).

If you think that'd help any, I'll see if I have enough bandwidth remaining this month to nab it (there should be enough).
 
No, I have not updated the drivers. They are Crimson 16.1.1, driver package 15.301.1811-160120a-300958C-HP (which seems to indicate a build time of Jan 20th, so fairly old).

If you think that'd help any, I'll see if I have enough bandwidth remaining this month to nab it (there should be enough).

Just grab a usb stick and grab it somewhere else , just wanted to warn you that desktop drivers are not always suited for APU.
 
Weeelllll, hell. I continually am running into this force-close wall, even when I've set the dang thing's Min/Max processor % at 99, which not only disabled Boost state but also drops it down a PState into P5 ([email protected]). However, the bit of silver lining to this is that even with the core speed significantly lower, the gaming performance in GRiD2 is no different, and may even be a little bit better since there's more thermal headroom for the GPU side.

Just for the helluvit I'll throw this old copy of Unigen Heaven on there and see if it ends up force-closing as well, or if it's just something to do with GRiD or not...

Some peculiar things worth noting:
- On some boots, the Boost is 2 multipliers higher on the P0 state for a max of 3.8GHz, yet the 2 other CPB states and base speeds are normal. Similarly, AMDMsrTweaker reports the max Core multi of 38, despite usually only being at 36.
- Even though I have Boost disabled, through AIDA64's sorta-hidden 'CPU-Tweakings' menu, the CPU still boosts like normal. Similarly, MSR readings reflect that Boost is "disabled". Yet I feel as though either this Home version of WIn10 isn't properly allowing for adjustments to the MSRs or this version of AMDMsrTweaker just doesn't support things properly enough, since despite changing things and the settings being reflected, nothing actually takes hold.
- I've looked in the Windows "Event Logs", to see what the reason for the "crash" of GRiD is, yet.... nothing appears. That's why I've now started to call it "force-close", and I have a possible theory as well... I'm beginning to wonder if maybe it is one of HP's programs on here that kick in when thermals reach a certain level? If running Heaven results in a close, then I may be on to something. I really am hoping that shutting down those programs/services will solve it, and that it isn't some highly embedded driver that is handling that... or worse the BIOS.

We shall see here shortly...


EDIT: Hmpf... So running Heaven didn't quite cut it, even full screen with everything cranked up and running for like 35minutes straight. I figured OK, it wasn't loading the whole APU fully, and fired up AIDA's stress test feature, running Heaven in a x720 Window... Still not enough. Temps are holding around 55C. Highest recorded shows to be 65C for the overall CPU and 71C for cores, but that dropped quickly and I'm pretty sure it was before the fan kicked on. Which I still don't know if that thing runs full tilt, given I have the entire bottom vent blocked right now and after 5mins it's only risen to just ticking over 60C, so it is quite heatsoaked.

However... There's quite a lot of throttling going on lol I don't know quite how it balances out in game, I'd have to be actively logging it in the background to know (and I"ll probably have to do that), but on a 2sec poll rate right now, the CPU is typically only running at.... 800MHz lol It averages about 1475MHz though.
GPU is not throttled as heavily, at least compared to that Dell and the A10-7300. Average GPU on this is roughly 559MHz (756 is full-speed).
Even the Northbridge takes a slight throttle, dropping from 1100MH to 1000MHz.

Starting to think maybe it IS GRiD2 related. That, or maybe memory related but... eehh... Dunno.
 
Last edited:
You might try getting a cooling pad with an usb fan and see if Grid2's crash to desktop issue goes away. Hopefully there are small vent slits in the bottom of the laptop case to help push cool air into the motherboard / processor area of the system.

I have the prior generation of AMD processor (A10-7300) laptop, and I found throttling issues related to heat during gaming were greatly reduced by a cooling pad with usb fan under the laptop.
 
I had one of those once... I think I gave it to my neighbor due to never using it lmao However... the bottom on this thing has barely any ventilation. <_> I don't even recall seeing any filter across the those vent holes to remove for better airflow; however, it may still help due to the aluminum body which conducts a fair amount of heat (though thankfully not enough to be uncomfortable on your lap, so props there, but the heatsink could've/should've been wider). S'pose now is a decent enough segway to show the interior shot that I may or may not have taken, when I may or may not have opened the laptop up... lol Which upon looking at the pics, damn I regret not taking a better/full shot :| It's just the board area, and doesn't show the gap next to it where the fan is, and all of the heatsink.
ENVY_Most-of-Inside.jpg ENVY_RAM.jpg


Check the event logs to see what is causing it to terminate, it should have some information in there :)
lol That was one of my bullet points:
- I've looked in the Windows "Event Logs", to see what the reason for the "crash" of GRiD is, yet.... nothing appears.

I had the Event Viewer open during play time, so after it force-closed I immediately refreshed the logs, but the only new entries were about the inability to access Windows Update since I have it shut off (so that I don't go over my bandwidth by downloading tons of Updates every few days).
 
I had one of those once... I think I gave it to my neighbor due to never using it lmao However... the bottom on this thing has barely any ventilation. <_> I don't even recall seeing any filter across the those vent holes to remove for better airflow; however, it may still help due to the aluminum body which conducts a fair amount of heat (though thankfully not enough to be uncomfortable on your lap, so props there, but the heatsink could've/should've been wider). S'pose now is a decent enough segway to show the interior shot that I may or may not have taken, when I may or may not have opened the laptop up... lol Which upon looking at the pics, damn I regret not taking a better/full shot :| It's just the board area, and doesn't show the gap next to it where the fan is, and all of the heatsink.
View attachment 4987 View attachment 4986



lol That was one of my bullet points:
- I've looked in the Windows "Event Logs", to see what the reason for the "crash" of GRiD is, yet.... nothing appears.

I had the Event Viewer open during play time, so after it force-closed I immediately refreshed the logs, but the only new entries were about the inability to access Windows Update since I have it shut off (so that I don't go over my bandwidth by downloading tons of Updates every few days).

Ahh, interesting, I had something similar with Forza 6 Apex, and it was the driver, but event log showed it
 
On Vacation Update: So I've had some actual play time with the system. Here are some of my observations that are specific to this device (and not quirks of Win10 or whatever).
Battery Life - While I haven't had too much unplugged time beyond the airplane, which also unfortunately wasn't much more than some music listening and *cough* (don't judge lmao) Scrabble Blast Deluxe playing. Still, it managed just fine IMO considering it's just a 4 cell. I would've been able to provide a much more accurate analysis as I was going to use the Delta In Flight network to watch videos (they now let you sort of do an On Demand service with the plane's media library, so if your seat doesn't have a built-in screen, you can use your laptop or mobile device to still watch stuff for free). However, thanks to Edge browser (and IE) being total derps, the system wasn't able to function because it thought Flash wasn't installed. Conveniently, they provide a download for it! Unfortunately, the Flash Installer won't exactly work with Win10 it seems :| Observe:
wtf chuck.png

Tablet Mode - I'll chalk this up to my own fault, albeit one I'm still uncertain as to what the cause is, but nevertheless... for whatever reason the screen and geometry sensors aren't functioning. For example, originally when you'd rotate the PC on its side, the screen would rotate with you (I think they call this "Reader Mode"). Also when you'd flip the screen a little past horizontal (with the base) it would switch into Tablet Mode (not Windows 10, but the system) which would thus disable the keyboard and touchpad. I'm trying to determined exactly what it was I changed/tweaked that could have caused both of these to stop functioning, but at the same time I can't really say with 100% certainty that it was me that is to blame :\

Keyboard - Just when I thought I was getting used to typing on it before I left for vacation, I end up using it more only to find that not much has changed really, and I can only chalk it up to the keyboard, no myself :( For example, typing certain keys are a huge pain as they don't always register. It's not for lack of pressing them either, nor is it that they're "failing" to detect the key (so not any sort of hardware failure), it's purely due to the keyboard's design. While typing that first sentence, I had to press the comma key 3 times for it to work after "changed really"... I even looked while I pressed it to make sure I wasn't on the corner or something, but no I was pretty much dead center, it's just crap. Also I can confirm that these do still use the scissor lift design on the keys, despite my thinking that it was something else. I had brushed off a couple pet hairs on the keyboard and for whatever reason the NumPad 6 key popped off, so I was able to inspect the design closely. All in all, I can only speculate that the reason why this keyboard 'sucks' compared to my old Toshiba, despite having nearly identical mechanical designing for how the keys function, is that this keyboard's rubber nipples are quite lot smaller than what my Toshiba had. I believe that the larger nipple would allow for two things: primarily that the key would 'depress' when your finger was at a more outside edge of the key, secondly that the increased contact pad's surface area would allow for a more accurate registering of presses. Not sure if that really makes sense, but I guess it sort of comes down to leverage and these smaller nipples prevent the conductive portion from being pressed against the pcb (either enough, or at all). Other than that, I notice that my ability to properly type is very "anal" with this keyboard when it comes to my positioning. It basically comes down to that if I'm not sitting with the laptop 'properly' situated in front of me like one might envision some illustration of "how to use a laptop" would depict, that my hand positionings are altered enough that cause vast typos due mainly to the keys not registering.

Touchpad - Some things I like, some things I don't :( I'm having more and more problems with it not detecting my tap-clicks, which is quite frustrating. I think I'm going to end up trying it setup with there being NO 'safe zone' boundary (to prevent accidental palm activation) as well as disabling the delay between keyboard usage and touchpad functionality. So often do I encounter where I just need to press one key and then having to wait a moment before the touchpad will start to function. Again, I understand the reasoning behind this, and this isn't so much an HP ENVY issue I don't think. The actual tap-click registering properly may indeed be hardware related, but I have no way of really knowing.

SOMETHING IMPORTANT TO NOTE! (but again, not an HP problem from what I can tell)
First of all let me start out by saying I've solved my Force-Close issue with that game!
After scouring the internet for why GRiD2 was just force-closing like that w/o any reason or log evidence, I came across a Steam thread where someone was experiencing something similar. One thing I HAD noticed was on one occasion, after the game exited suddenly, that Steam tried to launch it again. I thought maybe I had just tried opening it too many times initially, and so it finally was executing the oopsie after the first instance closed. Turns out someone else was having that same issue, but this time it was on a desktop and with NV graphics, persisting through driver changes and reinstalls of the game. Thankfully, someone commented on that thread and said to 'try this out, it worked for someone else'.
Well, I tried it also and... it fixed my problem! What I had to do was Stop the "Touch Keyboard and Handwriting Panel Service", which is easy enough via Task Manager. So whatever it is about that service, it seems to randomly an inexplicably cause either the system to think Alt+F4 has been pressed, or that Steam for whatever reason decides your game needs to be closed and restarted. Personally, I'm betting on the latter, given a 'phantom Alt+F4' wouldn't make Steam just so happen try to re-launch the game.

Either way, important thing is that solved the problem! What that 'means' for system usage? Not much, really. So far, all I've found that it does is kills the On Screen Keyboard; however, the icon in the task tray persists and pressing it will automatically restart the service. Furthermore, from what I could tell, Touchscreen funcationality (and touch pad for that matter) wasn't noticeably effected :) Overall I'd say it's a minor inconvenience, and so far has been unique to GRiD2 for me, even though my testing of Steam games has been limited to that, Besiege and Universe Sandbox.



On a related note... Seems that there are some issues with Catalyst drivers regarding Carrizo products... At least, the pre-DDR4 models (the 8000-series). I wasn't able to determine specifically what the 'bug' is or causes, but it made AMD decide to pull Carrizo support out of driver updates completely (well, I suspect just via INF) and even makes it a point to inform you that APU users need to use Catalyst 16.3.2, which is a number of months old (latest 'hotfix' was 16.7.2). The driver that the ENVY ships with says it's something between 15.12 and 16.1.1, but it's (I guess?) HP tweaked, so I don't rightly know. The driver version number says one thing, and depending on how AMD goes about that numbering system, it could even be some hybrid of 16.1 and 16.3.2 since the version claims to be the damn near the same as 16.3.2. I tried to update to that version using their lightweight web installer, and it said it'd upgrade the graphics drivers in addition to 2 or 3 other things, along with installing VulkanRT stuff... alas, after I installed it then rebooted, while the Vulkan stuff did appear to be added,, the driver still indicates as being the 16.1 "HP" version... So I seriously have no idea what's up.

Anyways, that's that so far :)
 
Hardware keyboard problems aside it still seems quirky and this is prolly not the first 2 in 1 product HP released.
For the browser problem I would say pick another :).

The version number reporting of the video card driver always been iffy , for the longest time people that updated drivers got reported their old version number in catalyst driver.
 
just a quick point the difference in ram speed per clock is about .0001sec, honestly just about the same. 1866@13 = .0139 2133@15=.0140
 
AMD shooting themselves in the foot again! I am curiously waiting for the desktop Bristol Ridge APU review, whenever that is going to be.
 
Back
Top