Where is Anandtech's review of Devil's Canyon??

[AtenRa]

What's wrong with them? They look fine to me.

Core i7 4770K lower than Core i3 ???

A10-6800K bellow 5800K ??

Core i7 4790K OC to 4.7GHz bellow Core i7 4770K default ???

 

[AtenRa]

If they really are to use such settings. Then they need to focus on whats relevant for it.

Smoother gameplay if you like

Yes i would like them to include FrameTimes

 

[Lepton87]

If they really are to use such settings. Then they need to focus on whats relevant for it.

Smoother gameplay if you like

How long is the benchmark? If it is 5 minutes long or something like that, that 100ms must be very significant

 

[ShintaiDK]

How long is the benchmark? If it is 5 minutes long or something like that, that 100ms must be very significant

1 minute I think. But I could be wrong.
http://techreport.com/review/26683/overclocking-the-core-i7-4790k/3

 

[SPBHM]

Although more games would be better to be included, non the less the data provided is excellent for 1080p because of the SLI results.

I don't think it's really relevant, it would be like using BF4 single player to compare CPUs... you need to dig deeper to find games where the CPU performance is relevant, and the most important thing, play the game to find the bad spots for CPU performance, also I'm always positive to exploring different settings to reduce GPU bottleneck on CPU comparisons.

the frame time variance explored by TR is also a good thing.

What's wrong with them? They look fine to me.

from the ones I posted

1- 4770K the slowest CPU
2- 6800K with the result of some HD 4600
3- 4770k faster than 4790 4.7GHz,
4- 4790K performance looks unrealistic compared to the 4770k

 

[Ryan Smith]

Core i7 4770K lower than Core i3 ???

A10-6800K bellow 5800K ??

Core i7 4790K OC to 4.7GHz bellow Core i7 4770K default ???

Hey guys, thank you for the reports. We integrated a ton of this stuff into Bench for this round, so we're admittedly still picking out errant typos. All of the noted problems have been fixed.

 

[AtenRa]

Hey guys, thank you for the reports. We integrated a ton of this stuff into Bench for this round, so we're admittedly still picking out errant typos. All of the noted problems have been fixed.

No problem, would you consider to include FrameTimes next time ???

 

[ShintaiDK]

Hey guys, thank you for the reports. We integrated a ton of this stuff into Bench for this round, so we're admittedly still picking out errant typos. All of the noted problems have been fixed.

Not all are fixed:
http://www.anandtech.com/show/8227/devils-canyon-review-intel-core-i7-4790k-and-i5-4690k/3

 

[AtenRa]

I don't think it's really relevant, it would be like using BF4 single player to compare CPUs... you need to dig deeper to find games where the CPU performance is relevant, and the most important thing, play the game to find the bad spots for CPU performance, also I'm always positive to exploring different settings to reduce GPU bottleneck on CPU comparisons.

the frame time variance explored by TR is also a good thing.

I dont believe it is relevant how you are going to play the game and what performance to expect if only you include spots where there are heavy CPU or GPU bottlenecks.

 

[cytg111]

just wanna say +1 for the reaction and the review is good stuff, thx

 

[cytg111]

Not all are fixed:
http://www.anandtech.com/show/8227/devils-canyon-review-intel-core-i7-4790k-and-i5-4690k/3

You know, like "Are you a cop?" - kind of question... "Are you getting paid to be pro intel" ? - all the friggin time? just curious, no harm no foul...

 

[Essence_of_War]

You know, like "Are you a cop?" - kind of question... "Are you getting paid to be pro intel" ? - all the friggin time? just curious, no harm no foul...

What?

 

[SPBHM]

I dont believe it is relevant how you are going to play the game and what performance to expect if only you include spots where there are heavy CPU or GPU bottlenecks.

it's a CPU comparison, it should focus on CPU dependent gaming, because that's why you buy a fast CPU if you are gaming, right? that's when it makes a difference, more CPU dependent games/scenarios are needed, or at least go into more detail and expose the frame times like TR is doing,

gaming performance can be difficult to be well represented on these short runs anyway

testing GPU limited gaming in a CPU comparison is almost like comparing, I don't know, ram speed, SSD speed?! something not really being affected to drastically by the CPU speed!?

not trying to sound to negative, because I enjoyed going trough the review, and I think the addition of the Dolphin emulator is quite nice when it comes to a kind of gaming CPU dependent and largely ignored by reviews.

 

[Durp]

Next time, spend your money, buy the hardware, spend a week and wright your own reviews the way you like them. Until then, you dont even have the right to talk about lazy or bias

10/10

it's a CPU comparison, it should focus on CPU dependent gaming, because that's why you buy a fast CPU if you are gaming, right? that's when it makes a difference, more CPU dependent games/scenarios are needed, or at least go into more detail and expose the frame times like TR is doing,

gaming performance can be difficult to be well represented on these short runs anyway

testing GPU limited gaming in a CPU comparison is almost like comparing, I don't know, ram speed, SSD speed?! something not really being affected to drastically by the CPU speed!?

not trying to sound to negative, because I enjoyed going trough the review, and I think the addition of the Dolphin emulator is quite nice when it comes to a kind of gaming CPU dependent and largely ignored by reviews.

I agree, as should anyone really. If you go back and look at the older CPU reviews, especially when Anand wrote them, the games were tested when the CPU was the limiting factor because that's THE ENTIRE POINT OF THE REVIEW! If the game wasn't CPU bound, he lowered the settings. If that wasn't enough, he lowered the resolution as well.

Kindly ignore those who cry about how unrealistic low resolution CPU benchmarking is. These guys are interested in hiding poor processor performance behind a GPU bound result. Most people aren't emotionally attached to a brand and would rather just see the truth. You can always include the usual "realistic" GPU bound result so those insecure few have a graph to post on various forums.

Save the GPU bound results for the GPU reviews? Also, as a few have already mentioned, please start joining the crowd by including frame times.

/constructive criticism.

 

[crashtech]

I'd like to see the test parameters carefully adjusted to give results that reflect as demanding a load as possible on both GPU and CPU, while not becoming so GPU bound that distinctions between CPUs can't be accurately determined. Overly low res. tests of CPUs introduce some questions as to whether a maximum CPU load is being imposed. Doing so would likely necessitate different quality and res. settings for each individual game, though.

 

[kawi6rr]

You know, like "Are you a cop?" - kind of question... "Are you getting paid to be pro intel" ? - all the friggin time? just curious, no harm no foul...

+1
Great review too

 

[DrMrLordX]

I think the point of the games review was to showcase what kind of GPU performance you get in the CPU, not to showcase how well the CPU can support an external piece of gaming hardware (namely, a dGPU). Now, realistically speaking, they SHOULD have had another section featuring as many CPUs as possible running powerful dGPUs to cater to the crowd that does such things. Such a review would have been later than the one provided for us here.

 

[Ryan Smith]

Not all are fixed:
http://www.anandtech.com/show/8227/devils-canyon-review-intel-core-i7-4790k-and-i5-4690k/3

You would have to be a bit more precise. There are a lot of graphs and a lot of individual results on that page.

You know, like "Are you a cop?" - kind of question... "Are you getting paid to be pro intel" ? - all the friggin time? just curious, no harm no foul...

We like to be as transparent as possible as far as getting paid is concerned. In short, no, we don't get paid by anyone to be pro anyone. We have a strict separation between our advertising partner and our editorial team. Ian and I do not know who will be running ads or what they are paying. Our paychecks are based on our body of work alone.

http://www.anandtech.com/home/about

AnandTech's business model is entirely advertising driven, which normally presents a clear conflict of interest within most similar organizations. I personally addressed this problem back in 1998 by completely removing advertising sales from AnandTech, Inc. We are represented by an exclusive third party who owns no share in AnandTech, Inc. (and vice versa). There is a strict separation between advertising and editorial. For the most part our editors aren't aware of who our current advertisers are at any given point in time. They typically find out when you do, by seeing an ad somewhere on a page.

 

[Homeles]

I think the review would benefit from including a power consumption chart, comparing the draw between DC systems and other configurations. Ian measured power consumption for Devil's Canyon, but it's difficult for readers to draw conclusions on efficiency without having to do a lot of extra work.

 

[StinkyPinky]

These cpu's live on the edge at stock. Currently the stock turbo boost from 4.0 is 4.4/4.3/4.2/4.2. I thought I would just bump it to 4.4 even (without change of voltage) and that was enough to blue screen under stress.

I notice the latest bios says "Improve Intel K-sku CPU performance" so that sounds interesting.

 

[Homeles]

These cpu's live on the edge at stock. Currently the stock turbo boost from 4.0 is 4.4/4.3/4.2/4.2. I thought I would just bump it to 4.4 even (without change of voltage) and that was enough to blue screen under stress.

I notice the latest bios says "Improve Intel K-sku CPU performance" so that sounds interesting.

Hope this doesn't come to bite Intel in the rear later.

 

[jpiniero]

Hope this doesn't come to bite Intel in the rear later.

You know, I was kind of wondering : is Intel guaranteeing that it can hit 4.4 for 3 years or 4?

 

[Homeles]

You know, I was kind of wondering : is Intel guaranteeing that it can hit 4.4 for 3 years or 4?

3.

 

[wilds]

I am stopping here at 45x on my 4690k. I am 'stable' at 1.26v, but I get higher scores on benchmarks at 1.276v without much of a temperature increase. On LinX, I am peaking at 91c at 4589 MHz.

If I go above 1.29v, temperatures are much higher, and throttling will occur. Using an EVO 212 cooler.

I was stable with 44x multiplier at 1.18v, but wanted to see how high I could go.
If I had better cooling, this chip could do 4.8-4.9 GHz no problem.

 

[jpiniero]

3.

I meant 4 Ghz.