I suspect that's because some examples tend to run over 95W?
From the Anandtech review.
If they will use the Wraith on the ZEN CPUs doesnt mean they will release 125W TDP SKUs.
A10-7870K comes with the D3 cooler and its only a 95W TDP CPU.
At 117W in our test, it seems a little over the 95W TDP listed, especially when the 7850K achieved 85W and under its 95W TDP. There will be variation within a SKU based on the binning quality, as I doubt the average 0.050 volts added to this SKU as I was told by AMD accounts for this much. Perhaps we ended up with a good 7850K sample but a mediocre 7870K sample. Alas with 100 CPUs we might get a better histogram of performance here, but we rarely get more than a single unit.
I suspect that's because some examples tend to run over 95W?
From the Anandtech review.
Running over 95W for limited periods of time does not affect TDP rating. From the same article, i7 4790K had a power delta of 110W even though its rated at 88W TDP, and i7 4765T had a delta of 56W even though it's rated at 35W TDP.I suspect that's because some examples tend to run over 95W?
Running over 95W for limited periods of time does not affect TDP rating. From the same article, i7 4790K had a power delta of 110W even though its rated at 88W TDP, and i7 4765T had a delta of 56W even though it's rated at 35W TDP.
Read my PS as well, even if we don't know the exact values and even if some efficiency adjustments were made, we still have a simple method of evaluating data through relative comparison: the 7870K is not the only data set with higher power delta than data sheet TDP, therefore we can simply bypass testing methodology and conclude data is not relevant for assessing chip TDP. Also, you don't have to accuse incompetence every time review data is not conclusive, there are more elegant and productive ways of communicating your findings.Read my post above, the reviewer has a flawed methodology, he use the delta at the main as power consumed by the APU, wich is just aknowledgment of incompetence to do such reviews...
Read my PS as well, even if we don't know the exact values and even if some efficiency adjustments were made, we still have a simple method of evaluating data through relative comparison: the 7870K is not the only data set with higher power delta than data sheet TDP, therefore we can simply bypass testing methodology and conclude data is not relevant for assessing chip TDP. Also, you don't have to accuse incompetence every time review data is not conclusive, there are more elegant and productive ways of communicating your findings.
59*C is not 13% warmer than 52*C.
Really wish they'd use better angles or at least flat out show that it doesn't obstruct the RAM slots if you were to use a kit with high raised fins.
Even that Corsair set, which isn't very tall, almost looks like it wouldn't fit, but of course the camera never goes far enough to show the clearance.
EDIT:
This image shows it will probably be fine with higher finned memory, but that's super tight for my liking.
http://images.bit-tech.net/content_...8370-cooler-review/amd-wraith-9-1280x1024.jpg
It is 13% warmer, relative to the freezing point of water.
the noise level and temperature difference to the old cooler is huge, it makes buying an aftermarket cooler much harder to justify.
the downside is that it seems to require a larger box, so I don't know how common it will be for new AMD CPUs to adopt this cooler and not one of the old smaller ones? this would be great for overclocking the cheaper CPUs.
It is not even a thing! How many % warmer is it relative to 52'C?
You get infinite...
How many % warmer is it relative to 53'C?
53-52=1'C
53-59=-6'C
-6/1*100%=-600% warmer? So it is cooler actually in relation to 53'C?
This is not how physics work.
For a proper scientific value go here:
http://forums.anandtech.com/showpost.php?p=38005144&postcount=9
LED lighting seems like a really unnecessary expense for AMD.
I suspect that's because some examples tend to run over 95W?
From the Anandtech review.
No, not surprising, but it would factor into the mix.
Zen might be fast enough to compete, but then a higher TDP could knock it back...
The 8370E's fan is twice the thickness of the 8320E's.yea the 8320e is 95W TDP and it has the simple aluminum cooler
But companies play games with TDP, too. Does the 4970K ever run at 88W for a strong workload?coercitiv said:Running over 95W for limited periods of time does not affect TDP rating.
Some numbers from the guru3D review:Does the 4970K ever run at 88W for a strong workload?
Interesting. The 8320E is rated for 95W and the 8370E is rated the same but the latter uses more power than the former.Some numbers from the guru3D review:
Normal usage scenario, no discrete GPU:
System idle power usage - 42W
System load power usage - 109W
Prime 95 load:
CPU Package Power: 86W
Max CPU Package Power: 112W
The answer to your question is: YES. Still, the chip TDP is 88W.
Zen wont have 256bit paths from the look of it. And it will hurt performance using it big time. My 6700K is a sub 65W chip not using it.
AMD's design philosophy is going to be to focus on the right engine for the right workload. So thats why they want to keep the Zen CPU FP at 128 bit. If somebody wants good FP performance AMD will have Zen APUs which crush Intel CPUs/SoCs for FP workloads.
As far as most productivity software and games are concerned integer performance is what matters and thats what sells CPUs in consumer market.
Correct me if I'm wrong...but isn't the plan with Zen to have 8 cores but 16 threads.
Afaik Intels' top dog 8core/16thread 5960X sits at 140W with 3Ghz base/3.5 ghz boost.
And isn't the 6700k rated at 91W for a 4core/8 thread?
I'm just saying....Intel may be massively leading in terms of performance...but their energy efficiency was never all that impressive in my eyes. To me it almost feels like all energy efficiency that Intel has comes mainly from being on a smaller die...and soon AMD will also be on a WAY smaller die.
I mean ffs they're jumping from 32 and 28 to 16 and 14.
Falling back into your old ways, I see
Considering AMDs current rep for hot/noisy/throttling with the stock cooler, this should eliminate that impression over time with Zen. Bonus points for likely being able to achieve a nice OC with the stock cooler.
My view is once people get their hands on high performance APU/SoC with stacked memory they will not look back. Anyone who wants serious FP performance will look at Intel Xeon Phi / Nvidia Pascal/ AMD Greenland and AMD Zen based APUs. I am looking forward to Zen APUs with HBM2 in both consumer and server segments.