Intel "Devils Canyon" Launch - Reviews, OC Info, Availability

[StinkyPinky]

We do have multiple employees, you know.

There will be a review of the 4790K. We just got the processor, so it will take some time.

If you guys only just got the CPU then that doesn't exactly shower me with hope that the rest of us will be getting one anytime soon.

 

[Burpo]

I like it.. Submitted by Nickshih@Asrock | 2014-06-13 10:00:33 :whiste:

http://valid.canardpc.com/ytgnpm

 

[StinkyPinky]

Apparently not due for retail until early july now

 

[Tweakin]

Pre-Order for June 25th at Newegg...

 

[Tsavo]

I think it just reflects Anand's changing tastes; he seems to find the mobile scene more interesting than PCs these days. Which is a shame for us.

Not to mention that makers of PC goodies haven't exactly been showering us with new toys to play with.

He's diversifying his coverage, which is a smart thing to do.

 

[notty22]

Amazon is shipping the Anniversary Pentium and 4970K on june 20th.
http://www.amazon.com/Intel-Core-i7-...ds=intel+4970k

 

[Homeles]

Not to mention that makers of PC goodies haven't exactly been showering us with new toys to play with.

He's diversifying his coverage, which is a smart thing to do.

No kidding. Desktop announcements have been dull all around.

 

[tidy]

.

 

[jj109]

How is it that such large improvements can be achieved by GPU drivers in CPU bound scenarios?

Well, the GPU drivers run on the CPU.

 

[tidy]

.

 

[BFG10K]

A good competent review: http://www.computerbase.de/2014-06/intel-core-i7-4790k-test-overclocking-kuehler-devils-canyon/

 

[jpiniero]

I thought that it was interesting that at stock the 4790K draws about 15W more at load than the 4770K does. Tom's review had the same thing, btw. It's much more than you would think given the 4W TDP difference even though it does run cooler.

 

[Atreidin]

I like it.. Submitted by Nickshih@Asrock | 2014-06-13 10:00:33 :whiste:

http://valid.canardpc.com/ytgnpm

I don't see VT-d in that screenshot but I see it here:
http://ark.intel.com/products/80807/Intel-Core-i7-4790K-Processor-8M-Cache-up-to-4_40-GHz

I hope it is just an issue with CPU-Z.

 

[tidy]

.

 

[ShintaiDK]

I don't see VT-d in that screenshot but I see it here:
http://ark.intel.com/products/80807/Intel-Core-i7-4790K-Processor-8M-Cache-up-to-4_40-GHz

I hope it is just an issue with CPU-Z.

Its an ES, not a retail chip either.

 

[Homeles]

That inefficiently? Inefficiently enough to displace the game by 20fps?

Easily, it just depends how bad off the original driver was.

 

[Lepton87]

On more than one occasion I've heard the folk wisdom that "Gamebryo likes cache", so that's my current hypothesis for Skyrim at least.

WoW behaves the same way it loves cache it's faster on 8 Core Xeon at 3.1 with 20mb cache then 3970X at 3.5GHz with 15mb of cache.

 

[frozentundra123456]

I thought that it was interesting that at stock the 4790K draws about 15W more at load than the 4770K does. Tom's review had the same thing, btw. It's much more than you would think given the 4W TDP difference even though it does run cooler.

Just goes to show that TDP is not the whole story. I am not sure what the full load turbo speeds are for the two chips, but probably about 12% faster for 4790k, so no way it will use only 5% more power.

 

[jalyst]

Thanks for responding Ryan. In fairness I think there is room for a little confusion due to Ian mentioning 10 days ago that testing should begin shortly. Maybe Intel took longer than expected to deliver the processor.

In case you're still reading, do you know if AT still plans to also offer a full Surface Pro 3 review?

Wanna know this too...
I also hate how they almost never respond when you ask about progress updates on these things.
It's not hard to spend 10 seconds responding, completely ignoring is pretty poor form.

 

[frozentundra123456]

A good competent review: http://www.computerbase.de/2014-06/intel-core-i7-4790k-test-overclocking-kuehler-devils-canyon/

Got excited at the 5.0ghz screenshot until I used google translate to see that it was on one core.

 

[BFG10K]

And again, we see the 6 core HEDT procs outperform the Haswells by a significant margin, even outperforming DC with it's high clocks, in Skyrim, a game that cannot take advantage of the extra cores. And it's even doing it at a 'real world' resolution.

This is really easy to test: either underclock your memory or enable XMP to boost it, then you can see the effect of memory performance. You could even drop to single channel mode.

If the game's performance still doesn't change much, it's probably cache bound. Some BIOSes even allow you to change the cache size so you can further test things that way.

 

[Homeles]

Wanna know this too...
I also hate how they almost never respond when you ask about progress updates on these things.
It's not hard to spend 10 seconds responding, completely ignoring is pretty poor form.

I'm fairly certain Ryan doesn't spend a terribly large amount of time browsing forums. I've found him to be very good at replying to emails, though.

 

[jalyst]

Yeah, he & some of the others are often okay, but it goes "into the black void" whenever it must be escalated to Anand.
And often it must be, because only he's in a position to answer Qns about *when* (roughly) or *if* something will be published.

 

[tidy]

.

 

[crashtech]

I've tried to simulate the performance of an i3 using an i7, but it's an imperfect test because cache size can't be altered on the fly... can it?