Intel Devil's Canyon: The Good, the Bad, the Ugly @ [H]

I wonder if that JF AMD guy that was on all the forums back around 2009 saying Bulldozer was going to set the world on fire, got hired on at Intel's marketing department. I kid.

Seriously though this is really disappointing and unlike Intel to make claims this far out of line with reality. I know Intel is a big ass global company but how can the left and right hand be this out of touch?
 
I got to give u some training
Click to expand...

Does the above quote remind anyone of Jobs' "Just avoid holding it in that way." (translation: You're holding the phone the wrong way.)? Apparently Kyle and Ryan are just overclocking the wrong way!!

Although I am surprised that the 4790K isn't giving at least a "toten" 100-200 MHz overclock advantage over a 4770K, I never thought it would be capable of overclocks of several hundreds MHz. If just changing the TIM and adding a few additional capacitors was all that was needed to add such increased overclock performance, Intel would have done this a long time ago!!

The reason why Intel charges the same amount for a 4790K as a 4770K is because Intel was always selling the 4770K as a under-performing part from the start. There was no reason why the 4770K couldn't have been a 4.0 GHz base clock CPU, especially since it was proven to overclock far past 4.0 GHz. Intel couldn't charge more for a 4790K because Intel isn't giving anything new or different with this CPU. Probably everyone who bought a 4770K also bought a Z motherboard and overclocked the 4770K way past 4.0 GHz right away, which makes the 4790K base clock inconsequential since the 4770K "real" base clock for the majority of users was more like 4.4 GHz or more.

It is great to see Intel with an actual focus on computer hardware enthusiasts and their needs. http://www.hardocp.com/article/2014/06/03/intel_devils_canyon_core_processor_presentation
Click to expand...

Perhaps Intel won't let us down with Haswell -E because they did with the 4790K!!
 
ChefJoe said:
The GPU manufacturers must be loving all the enthusiasts no longer needing to upgrade the CPU and having an extra GPU upgrade or two per motherboard cycle.
Click to expand...

+1

Still liking my new 4930k though. Definitely a different beast to overclock than my 3820 was.

I am not at all surprised by the outcome with DC though. When the biggest thing they had on the release was better TIM.. HAHAHAHAHA.
 
Entilza said:
I'll stick with me Dual Core 2 E8400, 6 years and still going strong :)

Waiting for Skylake
Click to expand...

You have no idea what you are missing with a newer setup. I can't stand to use Core 2 based systems anymore. Even the 1st gen Core i7 setups are pretty slow comparatively speaking.
 
That's like 500 euro staying in my pocket. Thanks for the heads up Kyle, as there is no point to upgrade my 2500k.

And if engineering samples need crazy voltage, then the retail versions will be even worse. No wonder the secrecy and lack of reviews at launch.
 
thanks for that! might be a decent upgrade from my 920 @ 3.8. (used to get 4.2, but its getting old. lol)

edit, meant to quote the person letting us X58 owners know about the xeons.
 
Cali3350 said:
Well, now im trying to decide whether or not to get a 4670k with a z97 and just wait till Broadwell (Broadwell works with z97 right?). Since I have a 920 the 4790k is still what id get but.. meh. Not sure its worth it over the 'old' i5 right now.
Click to expand...

I'd wait to make sure kinks if any are worked out of the Z97 and there gets to be enough reviews/feedback to weed out the good boards from the bad.
 
evilmedic said:
For people with x58 boards core i7 9xx platforms. Hex core westmere Xeon's (5600 series) are getting cheap on the second hand market and they also overclock well (5650,5660,5670).
Click to expand...

this


stilllllll waiting ... /talkingheads
 
I'm not, I've got hardware boredom lol.

My 2500K is 3.5 yrs old now, the longest I've ever used a CPU!
However, I found today that I can reduce my vcore and remain stable despite it being summer and pretty hot in the house.
I think its down to using different ram because I cant pin it on anything else.
So some excitement, gotta get yer kicks somehow.
 
My 2600K lives on :D Long live Sandy Bridge!!!


The fact that high ranking people(the president of Intel herself said so too) that Devil's Canyon would hit 5Ghz on air is just bizarre. Intel held out on sending review samples until after preorders went live so they could cash in while the hype was still there.

The launch of DC was suppose to breathe new life into the enthusiast market but this is just sticking the knife in a little deeper.
 
PGHammer said:
Irregardless
Click to expand...
No such word :p

A few games push my 2500K @ 4.5GHz very hard and Smooth Video Project does at times.
X3:TC especially could use a faster CPU, but it really needs orders of magnitude faster.
Still the more the better.
 
What bothers me the most about what h is reporting on DC OC is the part about using adaptive voltage. No thanks. Adaptive voltage mode causes stuttering and latency. I prefer manual 24/7. I'm at 3.5 or 4.6 always and I have very smooth web browsing for it. Adaptive doesn't work very well for me on haswell.
 
cyclone3d said:
You have no idea what you are missing with a newer setup. I can't stand to use Core 2 based systems anymore. Even the 1st gen Core i7 setups are pretty slow comparatively speaking.
Click to expand...

Sometimes, ignorance is bliss.

Can't wait to move to a first world country so I can play with the big boys. :D
 
Does this sentence make any sense to anyone???

"With the recent release of the Z97 Intel Express chipset, and Intel's communicating that it seemed a Z97 chipset motherboard would be required for Devil's Canyon usage, has surely left some with a bad taste in their mouth especially if you are an early adopter and geared up with a Z97 motherboard in hopes of getting that shiny new Devil's Canyon to 5GHz the moment you took delivery."

Why would one be disappointed with Devil Canyon if one already has a Z97 mobo?
 
I thought the DC and Haswell chips were exactly the same. The only difference was the thermal material between the chip and the heatspreader and possibly better binned chips for higher stock clocks.

When it comes down to it, what 4770k couldn't do 4.0 - 4.4ghz?

And then to top it all off, this is where I differ from most people. I was actually ecstatic that I could finally go direct to the cpu die again for cooling. Heat spreaders suck balls. Let me go direct die on a Haswell-E please. If I crack it my bad, if not I have the better cooled chip.

The only cpu's I owned that couldn't direct to die with a heatsink was Sandy Bridge 2600k and my Q9450. Every single other chip I have ever owned was either the bare die showing or I could remove the heat spreader.

Eh, DC is just a 4770 unless they changed anything on the chip itself, but I am unaware of those changes.
 
I will still be upgrading from my i3. So I won't complain about the higher base speeds.
 
Mostly Ugly... Why buy a new CPU when the one you have runs "fast enough" Intel needs to do a better job if they want any enthusiasts to follow them.
 
celeron300 said:
Mostly Ugly... Why buy a new CPU when the one you have runs "fast enough" Intel needs to do a better job if they want any enthusiasts to follow them.
Click to expand...

Well mine does not run "fast enough" (for my research - some operations I do take hours of all cores/threads at 100%) but they do not make a CPU that is enough of an improvement to make me want to switch from the i7 970 I purchased over 3 years ago for $365. DC was not even a candidate though. Haswell-E was but I expect the inability to improve performance at 22nm will yield the same result for Haswell-E.
 
Last edited:
young boy said:
Does this sentence make any sense to anyone???

"With the recent release of the Z97 Intel Express chipset, and Intel's communicating that it seemed a Z97 chipset motherboard would be required for Devil's Canyon usage, has surely left some with a bad taste in their mouth especially if you are an early adopter and geared up with a Z97 motherboard in hopes of getting that shiny new Devil's Canyon to 5GHz the moment you took delivery."

Why would one be disappointed with Devil Canyon if one already has a Z97 mobo?
Click to expand...

If you went ahead and upgraded your system to a new Z97 motherboard from a Z87 motherboard with hopes of installing a new Devil's Canyon and 5GHz, I would suggest that you would be disappointed for two reasons: 1. You did not need a Z97 motherboard for a DC processor. 2. It is highly unlikely that you will be hitting 5GHz with ease.
 
How disappointing. I was really looking forward to upgrading and having a good time OCing the 4790k.

Guess the good old i7 920 will keep on truckin' at 4.0 for another year. Heck, I don't even stress that old war horse all that much; upgrading was just going to be for fun and not really necessary.. Now that was a great quality CPU.
 
does DC seem to run any cooler than the 4770K? Did the TIM "upgrade" actually make any noticeable difference?
 
WtBadAss said:
does DC seem to run any cooler than the 4770K? Did the TIM "upgrade" actually make any noticeable difference?
Click to expand...


Yes and no. Yes, under stock voltages you can see the processor running somewhat cooler. No in that when we bumped the voltages up while overclocking, and the required voltages were higher, then it pretty much came in the same at 4770K when overclocking.
 
ChefJoe said:
The GPU manufacturers must be loving all the enthusiasts no longer needing to upgrade the CPU and having an extra GPU upgrade or two per motherboard cycle.
Click to expand...
I second that I just got a r9 270 third card for this x58 :)
 
WtBadAss said:
does DC seem to run any cooler than the 4770K? Did the TIM "upgrade" actually make any noticeable difference?
Click to expand...
Tom ran a 4770K and 4790K with the exact same multiplier and Vcore. They noticed a 6C difference under load.
 
Proxy said:
I wonder how well it would do with HT disabled.
Click to expand...

^^ This. Makes me wonder if that 5.5 number they were throwing around was with HT disabled and only a single core live. I'd be curious to know what motherboard they used to get that result. Had to have been a cherry picked chip.
 
mdswish said:
^^ This. Makes me wonder if that 5.5 number they were throwing around was with HT disabled and only a single core live. I'd be curious to know what motherboard they used to get that result. Had to have been a cherry picked chip.
Click to expand...

I would expect the LN2 helped quite a bit.
 
Now if only AMD could pick up the ball and run with it like they did when Intel fumbled with the P4. If that were to happen Intel might actually be motivated to push things forward more than just the minor improvements of late.
 
You say you have an engineering sample would it be more wise to talke about a retail unit before you start complaining about it?
 
Kyle - could you run all 4 cores at stock voltage 4.4Ghz? That is atleast still an attractive option for me.
 
Im sure when the original 5ghz announcement was made by the rep, intel engineers were shaking their heads ^_^.... just before the whips came out from management to hit that goal =P
 
My speculations about DC just got completely confirmed:

1. The claim by Intel that every DC could do 5+ GHz on air turns out to be a lie.
2. The claim by Intel that a new paste under the IHS was going to be "vastly improved" turns out to be a joke.

And I'm confident to say that using solder to alleviate #2 would have given truth to #1.
 
Spelling fixed. Thanks for the extra eyes. Can't believe I let that one get past me. DOH! - Kyle
 
Cali3350 said:
Kyle - could you run all 4 cores at stock voltage 4.4Ghz? That is at least still an attractive option for me.
Click to expand...

I did not try, but this worth a look and I still have the bench setup. I will give it a quick run. I was very much focused on 4.5GHz given that is our "stock" testing setup.
 
You must log in or register to reply here.
Back
Top