ASUS P5E3 Premium @ [H]

[quadnad]

Again the X48 is just an X38 with official support for the 1600MHz bus used by the QX9770. Otherwise it isn't any different. The Rampage Extreme and Formula are just like the Maximus Extreme and Formula boards except that they use a slightly different BIOS and the X48 north bridge chip in place of the X38 north bridge chip. So X48 isn't really all that exciting because it is just X38 re-hashed. You'd think that since the X48 chips are speed binned that they'd be better overclockers. My experience however finds that to be untrue. I've seen at least one X48 board that didn't overclock as well as its' X38 counterpart did. Though if you think about it virtually every X38 board can already do 1600MHz FSB (400x4) very easily. So really I doubt Intel is really even speed binning them. I think they just have sufficient confidence that they can officially sanction 1600MHz FSB speeds on that silicon.

So I suppose we won't be seeing a Rampage Formula review then.

 

[Dan_D]

So I suppose we won't be seeing a Rampage Formula review then.

You'll just have to wait and see.

 

[quadnad]

You'll just have to wait and see.

Good! Mainly because the Maximus was going to be my board of choice for my new build, but now that the X48s are arriving with the Q9450s, I may as well go with the Rampage.

Oh Dan, you and your ninja edits.

 

[Dan_D]

Good! Mainly because the Maximus was going to be my board of choice for my new build, but now that the X48s are arriving with the Q9450s, I may as well go with the Rampage.

Oh Dan, you and your ninja edits.

I am the NINJA EDIT MASTER!!!!!!!!

What ninja edit?

 

[quadnad]

I am the NINJA EDIT MASTER!!!!!!!!

What ninja edit?

heh
To be truly Ninja, you shoulda used #222222 for the color of that lettering! Like this!

 

[trinkus79]

No award? Given the raving review, I would have expected an [H] Gold award, or a silver at least!

Also, why does the board come with two wireless adapter antennas?

It's because it has 802.11n wireless. 802.11n uses multiple antenna's to get it's speed/range.

 

[thomasxstewart]

Ethernet has been weak point since windows 95 mostly you couldn't & that prevented other aspects from accepting any input, from there just forget about even changing page, it just got too mixed up..

Now with both MARVEL & REALTK controllers on one X48 Mainboard for gigabyte ethernet, this board stops CROSSTALK & CHATTER by requiring seperate drivers for each channel.

NEXT: HOW ABOUT FOUR Unique GIGABYTE ETHERNET CONTROLLERS, EACH FROM SEPERATE COMPANY, one for each core. WOWIE, maybe that get 2,3,4, sli, quad channel memory & WHOLE LOTTA oder stuff perking up with USB 3.0 final addon.

SignedHYSICIAN THOMAS STEWART VON DRASHEK M.D.

 

[Dan_D]

Ethernet has been weak point since windows 95 mostly you couldn't & that prevented other aspects from accepting any input, from there just forget about even changing page, it just got too mixed up..

Now with both MARVEL & REALTK controllers on one X48 Mainboard for gigabyte ethernet, this board stops CROSSTALK & CHATTER by requiring seperate drivers for each channel.

NEXT: HOW ABOUT FOUR Unique GIGABYTE ETHERNET CONTROLLERS, EACH FROM SEPERATE COMPANY, one for each core. WOWIE, maybe that get 2,3,4, sli, quad channel memory & WHOLE LOTTA oder stuff perking up with USB 3.0 final addon.

SignedHYSICIAN THOMAS STEWART VON DRASHEK M.D.

I do the testing all the time and I have to tell you that on all the boards I've tested NIC implementation is up to the manufacturer. Sometimes they are ok but usually they all kind of suck. The X38 and X48 motherboards are no exception to this rule.

 

[impaler]

NEXT: HOW ABOUT FOUR Unique GIGABYTE ETHERNET CONTROLLERS, EACH FROM SEPERATE COMPANY, one for each core. WOWIE,.

what would be the point?
even the best RAID HDDs can hardly break the 1Gbps barrier in normal operation
and anyway, how can you move 4Gbits of data around?
you would need massive networking hardware - that would cost hundreds of times more than the pc itself.

 

[Pepsiennis]

I do the testing all the time and I have to tell you that on all the boards I've tested NIC implementation is up to the manufacturer. Sometimes they are ok but usually they all kind of suck. The X38 and X48 motherboards are no exception to this rule.

I know you don't like me, Dan, but I trust your advice: if the onboard NIC implimentations are always lacking, which add-in card would you recommend? And is the difference worth the cost?

 

[Dan_D]

NEXT: HOW ABOUT FOUR Unique GIGABYTE ETHERNET CONTROLLERS, EACH FROM SEPERATE COMPANY, one for each core. WOWIE, maybe that get 2,3,4, sli, quad channel memory & WHOLE LOTTA oder stuff perking up with USB 3.0 final addon.

What the fuck would you need that for? Besides four network adapters wouldn't work that way.

SignedHYSICIAN THOMAS STEWART VON DRASHEK M.D.

I think you need to re-read the forum rules. You can find them here. http://www.hardforum.com/showthread.php?t=760666

You are in violation of rule #25.

(25) You are allowed one signature per post. A signature section is supplied for that usage. No one wants to see your name on your posts multiple times as it is annoying.

 

[thomasxstewart]

Wow, seems to be emotional problem here. I'd like to see why weak controllers on X48 tested so strong if its not two seperate drivers that don't cause disharmonious crosstalk on same channel by others being same , thus causeing BAD Chatter between two, Sort of like mentalator. btw POSTERS POST NAMES ALL TIME IN POST MESSAGE, YET THOSE NAMES ARE NOT CONSIDERED MULTIPLE AS ARN'T IN POSTERS NAME AREA. YOU IMPLY NOONE MAY USE MORE THAN ONE POST NAME FOR ENTIRE HARD OC FORUM, WHICH YOU VIOLATE..

SignedHYSICIAN THOMAS STEWART VON DRASHEK M.D.

 

[Caffeinated]

Kyle, I have always had the utmost respect for you. Enlisting Morry is tarnishing that image. I have no doubt he is thrilled with an Asus board; he is for hire, and it looks like he got paid by Asus.

I won't tell you how to run your website, but don't front a shill and expect me to visit. You are as hard a member of this site as most of your core members; your endorsement of him leaves me puzzled. You earned a WTF?

 

[PanzerKrieg13]

@ thomasxstewart ;

Ok, seriously, you're annoying. Not only is your caps lock hurting my eyes, but when combined with your horrible grammar, your posts are about the equivalent of having your hand cut off with a rusty razor blade, and then proceeding to shove your bleeding arm into a bucket of salt.

Just my $0.02,
- Joey

 

[Caffeinated]

@ thomasxstewart ;

Ok, seriously, you're annoying. Not only is your caps lock hurting my eyes, but when combined with your horrible grammar, your posts are about the equivalent of having your hand cut off with a rusty razor blade, and then proceeding to shove your bleeding arm into a bucket of salt.

Just my $0.02,
- Joey

Got to agree with you there.

 

[Dan_D]

Wow, seems to be emotional problem here. I'd like to see why weak controllers on X48 tested so strong if its not two seperate drivers that don't cause disharmonious crosstalk on same channel by others being same , thus causeing BAD Chatter between two, Sort of like mentalator.

No. You clearly do not have an understand of how the PCI Express or PCI busses work and you do not have a basic grasp of the Windows driver model.

btw POSTERS POST NAMES ALL TIME IN POST MESSAGE, YET THOSE NAMES ARE NOT CONSIDERED MULTIPLE AS ARN'T IN POSTERS NAME AREA. YOU IMPLY NOONE MAY USE MORE THAN ONE POST NAME FOR ENTIRE HARD OC FORUM, WHICH YOU VIOLATE..

SignedHYSICIAN THOMAS STEWART VON DRASHEK M.D.

No. There is a signature section in each of your posts. You can sign your posts in that section. That is what it is there for. Plus I can see your name at the top of your posts. There is no reason to post your fucking name again at the bottom of each post you make. It is irritating. It is irritating like your usage of the caps lock key which is similar to fingernails on chalk boards or that sickening smell in a morgue that never goes away even years after the morgue has ceased being used.

 

[Zoran]

I suppose I should be happy then with my Abit IX38QuadGT running a Q6600 @ 8x513 with 2GB of OCZ 800MHz DDR2 @ 1026MHz 5-4-5-13 2T getting the same ~8,5GBps mem bandwidth @ half the clock with less heat....

...Till I'll desperately NEED 3 3870X2-s


BTW: I understand you're time limited, and please don't take me wrong - I like your reviews - But I think you guys should really use Quads instead of DC-s for overclocking tests... or do both. I've seen boards to do almost 600(2400)MHz FSB with a Dual and not even coming close to 500 with a Quad. Also you should consider filling up all the mem slots when doing overclockin trials... And maybe you could describe your overclocking more in-depth, and maybe try a little [H]arder... Or else I don't see why this site is called [H]ard OC P...

 

[Dan_D]

I suppose I should be happy then with my Abit IX38QuadGT running a Q6600 @ 8x513 with 2GB of OCZ 800MHz DDR2 @ 1026MHz 5-4-5-13 2T getting the same ~8,5GBps mem bandwidth @ half the clock with less heat....

...Till I'll desperately NEED 3 3870X2-s


BTW: I understand you're time limited, and please don't take me wrong - I like your reviews - But I think you guys should really use Quads instead of DC-s for overclocking tests... or do both. I've seen boards to do almost 600(2400)MHz FSB with a Dual and not even coming close to 500 with a Quad. Also you should consider filling up all the mem slots when doing overclockin trials... And maybe you could describe your overclocking more in-depth, and maybe try a little [H]arder... Or else I don't see why this site is called [H]ard OC P...

I can understand where you are coming from on the quad core processors. No doubt we'll switch to quad core processors at some point, but that's up to Kyle. As for the 4 DIMM testing, I can already tell you that we tried that, and it was a pain in the ass. Most boards out today give us a ton of issues in a 4x1GB confiugration.

On some reviews I've tried it and it worked fine. But I can count on one hand all the boards that did it well and I can't easily count the ones that don't.

On the subject of overclocking, we try plenty hard enough to get the highest overclocks. However we are only interested in stable overclocks and nothing else. We also don't use cherry picked motherboards. We use retail boards most of the time. In over two years I've only reviewed two motherboards that weren't retail samples. The DFI Xpress 200 chipset board and the Intel D5400XS covered in the Skulltrail article. Just some of the reasons why our overclocks don't always match what some other sites report.

 

[Zoran]

I understand that you don't care about anthing else but stable OCs! Who does, anyway [except those crazy enough to play with liquid/solid gases]?

But you could at least describe what stages you went through, what settings you've tried and what was the situation throughout the testing... Voltages/percentages/Cooling modifications/etc. Used for the highest stable OC would be appreciated I think.

Information about the settings you've used could be useful for one's own conclusions. I understand that every single piece of HW behaves completely different, but you can find similarities and/or theese infos could help people to solve their own issues they can't find the cure for...

As for the Quads: I would suggest you to shift as quickly as possible as quads become more and more popular and the results achieved with Duos seem to be completely irrelevant/useless if you build a system with a Quad.

I personally had to change three times[boards] till i got the described results. I've had an IP35Pro that could do 486MHz stable [unfortunately it had power circuitry problems], then a P5E which was one of the worst boards ever i'd gave money for [straight after the strikerextreme], and now the IX38QuadGT which is EXCELLENT in all ways despite the amount of negative press it received.

About the RAMs: see that's exatly the kind of info I was curious about... There's far too little to no info about stability and/or usability of these "inexpensive" boards with all expansion slots populated on the net...

I know that the more HW you put in the system, the more time it takes to do the actual testing, but hey, isn't that the reason you're here for? I think many would be glad to find this kind of information as it is practically "non existent" on the net...

At least you could say whether the board would accept 4x2GB low latency DDR2 800/1066DIMMs If nothing else..

You guys do reviews completely different to others [that's why so many love/hate You], so why don't you provide data "noone" else does...

 

[TriPpeD]

"The board is available at ZipZoomFly for $344.99, although they have sold out for the time being."

Hard to believe they are selling this board as Asus ain't even released it to the consumer,sure it's the same board?
Anyone have a reliable date as to when this will be available, some say mid Feb other say mid March.

“Critics are like eunuchs in a harem; they know how it's done, they've seen it done every day, but they're unable to do it themselves.” BB

 

[TriPpeD]

From AnandTech http://www.anandtech.com/weblog/showpost.aspx?i=377

Intel X48 -

According to several motherboard manufacturers, it now appears that final X48 silicon will not reach the suppliers until early to mid-March. All indications are that X48 products will not be in the retail market until the later part of March now. This is an abrupt change from the expected release of retail product later this month. However, considering the number of dates we have been given on this program, it's not surprising we have yet another schedule change. In their defense, Intel has just said Q1 as of late, but we know from Internal documents that Q1 originally meant January. Oh well, it gives us a little more time to play with NVIDIA's upcoming 790i besides reporting on some interesting P35 and X38 boards coming to market at the lower end of the price scale.
Well i guess that settles it, a whole month! ahhh

 

[RussianHAXOR]

http://www.tigerdirect.com/applicat...55-2371&SRCCODE=GOOGLEBASE&CMP=OTC-GOOGLEBASE

This video proves it.

 

[FrgMstr]

Kyle, I have always had the utmost respect for you. Enlisting Morry is tarnishing that image. I have no doubt he is thrilled with an Asus board; he is for hire, and it looks like he got paid by Asus.

I won't tell you how to run your website, but don't front a shill and expect me to visit. You are as hard a member of this site as most of your core members; your endorsement of him leaves me puzzled. You earned a WTF?

I guess our image has been tarnished for many years now. It is not like Morry just got here.

You know sometimes, not everyone is going to have the same opinion as you. It is sad that you are so closed minded that it makes someone with a viewpoint other than your own a shill.

Morry is not going anywhere, so I guess that means you will not be visiting any more. Don't let the door hit you in the ass.

Consider this your personal invitation from me to leave the free services behind that we provide you and never return. Now put your money where your keyboard is GTFO. I don't need your baseless made up accusations about our staff and won't allow it. I can take criticism all day long. But you have crossed the line with your accusations and I will not allow it.

 

[Dan_D]

http://www.tigerdirect.com/applicat...55-2371&SRCCODE=GOOGLEBASE&CMP=OTC-GOOGLEBASE

This video proves it.

They are definitely similar.

 

[FrgMstr]

They are definitely similar.

You can do that with most asus motherboards.....the last three asus motherboards I have put in my case did not even require me to rework any of my cabling. Everything is in the same spot most of the time. And quite frankly, if you have a solid PCB, there is little reason to change it as long as the cost ban be rolled into the production product.

So as far as "proving" anything...it proves they have the same layout.

Read the reviews and forums, find a stable mobo that is supplying the features and looks to have the overclocking you want, and buy it. Don't purchase it because it has a fancy name on it.

 

[FrgMstr]

Wow, seems to be emotional problem here. I'd like to see why weak controllers on X48 tested so strong if its not two seperate drivers that don't cause disharmonious crosstalk on same channel by others being same , thus causeing BAD Chatter between two, Sort of like mentalator. btw POSTERS POST NAMES ALL TIME IN POST MESSAGE, YET THOSE NAMES ARE NOT CONSIDERED MULTIPLE AS ARN'T IN POSTERS NAME AREA. YOU IMPLY NOONE MAY USE MORE THAN ONE POST NAME FOR ENTIRE HARD OC FORUM, WHICH YOU VIOLATE..

SignedHYSICIAN THOMAS STEWART VON DRASHEK M.D.

You were warned nicely, and now you are gone. I don't think many here will miss you.

 

[HotBBQ]

AnandTech's review of the board just popped up today. They seemed to be falling over themselves with praise for the board. If the price is indeed $299 I think I will be snatching one of these up as soon as they hit the e-tailers.

 

[KCComp]

It appears [H] did not have a final version of the board to reiew.

Here is the [H] reviewed version:

And here is the version on Newegg (also shown in Anand's review):

I can see 3 major changes right off. First, the final board is missing the top PCI-E x1 slot, which is a deal-breaker for me, because that's the only place in my configuration I can plug in a NIC (both onboard NICs are trash when used at gigabit speeds, I had to go to an Intel PCI-E x1 NIC).

Secondly, and probably more important to the enthusiat community, is that the vDIMM regulation is completely different. I'm not enough of a power-component expert to tell you whether it's better or worse, design-wise.

Third, the north bridge voltage regulation is different, there appear to be 2 regulators on the final version instead of one on the [H]-reviewed version. This extra regulator may have necessitated the removal of the PCI-E x1 slot (it's hard to see in the linked Newegg pic, but if you look at their other angled shot you can see the 2 regulators).

 

[HotBBQ]

I can see 3 major changes right off. First, the final board is missing the top PCI-E x1 slot, which is a deal-breaker for me, because that's the only place in my configuration I can plug in a NIC (both onboard NICs are trash when used at gigabit speeds, I had to go to an Intel PCI-E x1 NIC).

Secondly, and probably more important to the enthusiat community, is that the vDIMM regulation is completely different. I'm not enough of a power-component expert to tell you whether it's better or worse, design-wise.

Third, the north bridge voltage regulation is different, there appear to be 2 regulators on the final version instead of one on the [H]-reviewed version. This extra regulator may have necessitated the removal of the PCI-E x1 slot (it's hard to see in the linked Newegg pic, but if you look at their other angled shot you can see the 2 regulators).

Can you put your NIC card in the black PCIe slot? Seeing as it is a mechanical x16 slot, but only supports x1 and x4 lanes it should be OK, but I'm not positive.

As far as the power stuff goes I pulled this off of their website and the [H] review respectively. It looks like the released board has more vChipset steps (49 compared to 34). That may account for the changes you see. I didn't find anything different about the vDIMM stuff, but I don't know what to look for.

ASUS Website
- vDIMM: 65 -step DRAM voltage control
- vCore: Adjustable CPU voltage at 0.00625V increment
- vChipset (N.B.): 49-step voltage control
- vFSB Termination: 16-step reference voltage control
- vCPU PLL: 65-step CPU PLL voltage control

Reviewed Board Specs
- vDIMM: 65 -step DRAM voltage control
- vCore: Adjustable CPU voltage at 0.00625V increment
- vChipset (N.B.): 34-step voltage control
- vFSB Termination: 16-step reference voltage control
- vCPU PLL: 65-step CPU PLL voltage control

I just purchased this board and it should be here by Thursday. If I'm lucky ShopBLT will have my Q9450 here before the weekend and I can start my new build.

 

[JCNiest5]

Interesting about the two versions. I will be able to tell which one is mine on Friday when mine arrives.

 

[sauria]

I just got mine. I got no remote control in the package, no PCI-E x1 slot. Rev 2.00G

 

[HotBBQ]

I just got mine. I got no remote control in the package, no PCI-E x1 slot. Rev 2.00G

Ditto.

 

[sauria]

I just built a P5E3 Premium with Windows XP SP 2 (loaded ICH9 drivers from floppy), e8400, 8800GT, Pioneer SATA Blu-Ray BDC-2202B, Barracuda 7200.11. I had to install the OS with IDE mode to boot from the DVD drive. I have no Parallel devices and a floppy drive. Is there Any way to change to AHCI? I get BSOD if I try to change it back in the BIOS. Thanks. DO I need to install on an IDE DVD and then install the new SATA DVD?

 

[sauria]

Thanks HotBBQ -- how the build going? I'm frustrated with the AHCI thing.

 

[HotBBQ]

Thanks HotBBQ -- how the build going? I'm frustrated with the AHCI thing.

It went fairly smoothly. I decided to go with Vista Home Premium w/SP1. I didn't have any problems with my SATA DVD drive or with my RAID 0 array. It all worked without installing any extra drivers. I had a little trouble with my X-Fi and drivers, but go figure, it is a Sound Blaster. I am looking forward to OC it, but I am new to the OC stuff and I want to go slowly and learn what to do.

 

[sauria]

Great HotBBQ - I guess Vista doesn't need a special AHCI driver?

 

[HotBBQ]

Doesn't appear so. BTW, it is a Samsung SH-S203N drive.

 

[sauria]

No, Pioneer SATA Blu-Ray BDC-2202B. Is that your drive?

 

[HotBBQ]

No, Pioneer SATA Blu-Ray BDC-2202B. Is that your drive?

I think you read 'is it' instead of 'it is'. In any case, mine is the Samsung drive.

 

[sauria]

Yeah, I'm tired -- working on trying to reinstall windows in the AHCI mode on this board. Thanks