• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD Ryzen (Summit Ridge) Benchmarks Thread (use new thread)

Page 188 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
bjt2 said:
There are probabily platform and VRM limits on 1151 MB... So this is the maximum allowed by the safety margins... INTEL does not want a CPU that does not work on some MB...
Click to expand...
Looks like the most minimal increase possible with plenty of room if push comes to shove.

Intel is extremely confident.

Sent from HTC 10
(Opinions are own)
 
KTE said:
Looks like the most minimal increase possible with plenty of room if push comes to shove.

Intel is extremely confident.

Sent from HTC 10
(Opinions are own)
Click to expand...

I dunno, putting HT on an i5 (even if the price is a bit more than the 7600K) is a pretty big deal. They were planning on doing this with Coffee Lake but doing it now seems a bit much.
 
@ltc8k

It's very simple : money!

Let's say a die costs 50$ - so you make more money selling them for 500$ than for 200$ - because if that u take the good binned dies with higher frequencies for the top end 8c SKU and the low clocking parts for the cheaper 4c
 
Zucker2k said:
What about the gist on quadcore ryzen debuting at 3.2ghz? What the actual......, if true, that is.
Click to expand...
It wouldn't be that bad.

i5 7400 - 3GHz base
i5 7500 - 3,4GHz base
i5 7600 - 3,5GHz base
i5 7600K - 3,8GHz base

All Ryzen's are also unlocked, on the other site there's only the 7600K as an i5 which you can OC.

Edit: My assumption is that AMD will compete (pricewise) with a 4C/4T Ryzen against i3, a 4C/8T Ryzen against i5 and so on...
If that's the case frequency shouldn't be a big deal.
 
KTE said:
Looks like the most minimal increase possible with plenty of room if push comes to shove.

Intel is extremely confident.

Sent from HTC 10
(Opinions are worthless)
Click to expand...

Where is their plenty of room within the warranty TDP limits of their sockets/boards?

For instance, we known that the current i7-7700K has a thermal limit of at most 130W[1].

So at most Intel can raise the clock by 300 MHz[2]. They've already ate 1/3rd of that with this current rush job.

Furthermore, they will also have to have a bit of a safety tolerance, so chances are they'll not push beyond 200 MHz jump on the i7-7700K. Which means in reality they've ate half of their cake.






[1]https://ark.intel.com/products/97129/Intel-Core-i7-7700K-Processor-8M-Cache-up-to-4_50-GHz

[2]http://www.anandtech.com/show/10968...mpion/11?_ga=1.152659326.698670777.1483630983
 
Lovec1990 said:
^^ why are AMD stocks rising so fast it is becouse of Ryzen if it is how does unreleased and untested CPU rises stocks so high?
Click to expand...

There are trillions of dollars of Fed funny money floating around desperately searching for any kind of yield. If the wind blows in a company's direction, their stock will get bought. Look at AMZN's P/E... With not even a single passmark baseline uploaded, all we can say for sure is that it is pure speculation.
 
sm625 said:
With not even a single passmark baseline uploaded, all we can say for sure is that it is pure speculation.
Click to expand...

Lack of leaks is not equal to lack of performance, at least not in this case.

I dunno why people are panicking about QC Ryzen's clocks? ALl AMD needs is good ST Turbo(4Ghz or a bit more) and base can be low(ish) ~3.5Ghz due to the fact it will support SMT in the price range where it will compete with KL i5s. So 3.5Ghz x ~1.25x (SMT yield) =4.2Ghz. Voila, close to rumored 7640K level if it lacks SMT, for less $.
 
Guys, if you think about this. 100W CPU, 4.3 GHz, base clock, with 4.6 GHz boost? 4C/8T Core i5 to fit in that range?

Two reasons. 4C/8T actually has pretty damn high core clocks, at lower power, and cost than Intel. And also 6C/12T, with not so high core clocks, with similar TDP, and similar cost to Intel.
 
They solution to all potential problems is good ST performance.

Glo.: we have very little and even that, only PR to go off.

One is forced to use history, economics and rationale.

Sent from HTC 10
(Opinions are own)
 
One of the (minor) details about XFR that I recall from AMD slides is that it moves clockspeed in 25 MHz increments.

My guess is it does 2d algorithmic analysis of the chip's clockspeed based on temperature and voltage. It will likely do in real-time what we overclockers do over hours of painstaking testing and tuning: observing temp vs. voltage and adjusting clockspeed (and voltage) upward where possible until temp + voltage thresholds are met. Then it backs down. It could be a little complicated because it will be a series of formulae that vary in two variables, not just one.

Only thing I can't figure out is: how will XFR know when failure due to high clock/low voltage is imminent? Unless it has a preprogrammed table of clockspeed/voltage combos (which it might), it will never be able to run the razor's edge. Traditional overclocking requires you to push your chip until it starts to fail, and then back clocks off from that point. That is, if you want the most out of your CPU. I'm not sure how the XFR system would be able to do that without bringing down the whole house of cards.
 
KTE said:
They solution to all potential problems is good ST performance.

Glo.: we have very little and even that, only PR to go off.

One is forced to use history, economics and rationale.

Sent from HTC 10
(Opinions are own)
Click to expand...

So you believe Ryzen is another bulldozer and will have bad ST performance?
 
Atari2600 said:
Quasi-final?

They were out running around 6 months ago for motherboard verification.

In no way, shape or form are they final!
Click to expand...
6 months ago they had 3Ghz base clocks or less, keep up with your leaks. So yes, 3.2Ghz base clock is higher than any clock on any 4c ES we have heard about until today. That actually makes it sort of a very big deal.
Glo. said:
Two reasons. 4C/8T actually has pretty damn high core clocks, at lower power, and cost than Intel. And also 6C/12T, with not so high core clocks, with similar TDP, and similar cost to Intel.
Click to expand...
Sometimes reading your posts is funny.
 
It seems like AMD isn't trying to pin 4c/8t vs 4c/8t perhaps this is what the 6c/12t is being targeted at and is eating up all the higher clock salvaged dies leaving whats left for the 4c/8t SKUs.
 
DrMrLordX said:
One of the (minor) details about XFR that I recall from AMD slides is that it moves clockspeed in 25 MHz increments.

My guess is it does 2d algorithmic analysis of the chip's clockspeed based on temperature and voltage. It will likely do in real-time what we overclockers do over hours of painstaking testing and tuning: observing temp vs. voltage and adjusting clockspeed (and voltage) upward where possible until temp + voltage thresholds are met. Then it backs down. It could be a little complicated because it will be a series of formulae that vary in two variables, not just one.

Only thing I can't figure out is: how will XFR know when failure due to high clock/low voltage is imminent? Unless it has a preprogrammed table of clockspeed/voltage combos (which it might), it will never be able to run the razor's edge. Traditional overclocking requires you to push your chip until it starts to fail, and then back clocks off from that point. That is, if you want the most out of your CPU. I'm not sure how the XFR system would be able to do that without bringing down the whole house of cards.
Click to expand...


I have found a link to a PDF explaining the AVFS, behind a paywall, on IEEExplorer. Since i am a researcher, my organization has access to this article and i read it.

In short, for Carrizo (and almost surely for all subsequent products) there are 10 units spread across the chips each of which have 50 replica of critical path circuits (for a total of 500) and with a complex circuit collect statistics on its delay at various Vcore, to calculate Fmax at those Vcores. This Statistic is calculated on demand by microcode or SMU and allow to precisely know what Vcore apply to a certain frequency. Obviously in Zen this will be useful also to calculate the Fmax, given the current situation (temperature, chip age, and so on)...
 
Lovec1990 said:
^^ why are AMD stocks rising so fast it is becouse of Ryzen if it is how does unreleased and untested CPU rises stocks so high?
Click to expand...
If AMD has a great server chip*, and if yields are good, and if the market accepts AMD server chips, and if Intel drops the ball when they release their next server chips this summer, then AMD has huge potential gain. The server market is massive and AMD has been cut down to a lean company without much excess cost. If all of those are true, then AMD could easily be worth $25/share (or more). But, that is a lot of "ifs".

AMD always shoots up before a launch. Most of the time, if falls back down after a short period. Either the AMD launch wasn't as hyped, or Intel responded (often by putting in one more top processor and knocking all other processors down one price range). I think AMD isn't $25/share now due to this risk.


* I speak of server chips because Ryzen seems to be geared for that at launch.
 
Status
Not open for further replies.

TRENDING THREADS

Back
Top