AMD's Future Chips & SoC's: News, Info & Rumours.

[goldstone77]

Compression tends to be a more serial based operation, and is a good test for single core performance.

Decompression tends to be a more parallel based operation, and is a good test for total CPU performance.

Good is subjective term. And based on other synthetic benchmarks in comparison is only shows those processors are more or less turned for each operation.

 

[goldstone77]

Heh, Linus always makes my day when he is angry about something.

In other news: http://www.tomshardware.com/news/amd-raven-ridge-apu-specifications,36375.html

Is there anything in there that we didn't already know?

XFR for the top model seems to be 4.2GHz. Apparently the memory controller can be overclocked up to DDR4-3600 stable.

There is an interesting speculation that says those chips are already using the 14nm+ process.

EDIT: Not just speculation. AMD James confirms

https://www.reddit.com/r/Amd/comments/7s56pd/the_upcoming_ryzen_apus_seem_to_already_be_the/dt2kxpt/

If you watch the video he says these are similar with desktop in the way they deliver power to the CPU, precision boost higher frequencies at lower voltages, but the desktop CPUs are 12nm. We have some sample going out to partners they are not full performance or final product yet.
https://youtu.be/7MPxN6MEgbk?t=400

 

[juanrga]

Heh, Linus always makes my day when he is angry about something.

In other news: http://www.tomshardware.com/news/amd-raven-ridge-apu-specifications,36375.html

Is there anything in there that we didn't already know?

XFR for the top model seems to be 4.2GHz. Apparently the memory controller can be overclocked up to DDR4-3600 stable.

There is an interesting speculation that says those chips are already using the 14nm+ process.

EDIT: Not just speculation. AMD James confirms

https://www.reddit.com/r/Amd/comments/7s56pd/the_upcoming_ryzen_apus_seem_to_already_be_the/dt2kxpt/

If you watch the video he says these are similar with desktop in the way they deliver power to the CPU, precision boost higher frequencies at lower voltages, but the desktop CPUs are 12nm. We have some sample going out to partners they are not full performance or final product yet.
https://youtu.be/7MPxN6MEgbk?t=400

12nm is a marketing name for 14nm+.

 

[YoAndy]

Ryzen 2 chips are due to debut in April So instead of a 3.2GHz base clock and boost to 3.6GHz as seen on the Ryzen 5 1600, the new 2600 looks set to lift the base clock to 3.4GHz and boost to 3.8GHz.

 

[goldstone77]

I find it most curious that a week before the 2600 engineering sample was seen on sisoftware we had the leak of an i5-8500@~$199(guess). 3.0GHz base 4.2GHz(6C) and 4.3GHz(1C) boost. Also, with the 2600 it appears to be inline with 10% estimates by GloFo, and looks like Intel is adding segmentations to answer Ryzen products. It will be interesting to see how high they can tweak 2000 series on the top end. 4.4GHz(8C)4.5GHz(1C) for the best case scenario.

 

[YoAndy]

It would be interesting to see a 4.5 but I'm not sure about that.

 

[YoAndy]

I find it most curious that a week before the 2600 engineering sample was seen on sisoftware we had the leak of an i5-8500@~$199(guess). 3.0GHz base 4.2GHz(6C) and 4.3GHz(1C) boost. Also, with the 2600 it appears to be inline with 10% estimates by GloFo, and looks like Intel is adding segmentations to answer Ryzen products. It will be interesting to see how high they can tweak 2000 series on the top end. 4.4GHz(8C)4.5GHz(1C) for the best case scenario.

So the 8400 will have no point.

 

[goldstone77]

I find it most curious that a week before the 2600 engineering sample was seen on sisoftware we had the leak of an i5-8500@~$199(guess). 3.0GHz base 4.2GHz(6C) and 4.3GHz(1C) boost. Also, with the 2600 it appears to be inline with 10% estimates by GloFo, and looks like Intel is adding segmentations to answer Ryzen products. It will be interesting to see how high they can tweak 2000 series on the top end. 4.4GHz(8C)4.5GHz(1C) for the best case scenario.

So the 8400 will have no point.

Well, it's still a little cheaper and will probably be really close in performance. I think Intel wants to have a chip to compete against every price bracket Ryzen would have something to offer so just to have a counter. They could even drop the 8400 down a few more dollars and really put a hurting on Ryzen. Think of the 8400 around $168 or so... That would put a hurting on Ryzen. Not like they are doing all that well with Intel holding ~90% market.
i5-8400
Recommended Customer Price
$182.00 - $187.00

 

[-Fran-]

Well, considering that 100Mhz steps in their lineup will not be massive jumps in performance, I'd say the 8400 will continue being the best P/P from the Covfefe Lake lineup until they phase it out or churn out new stuff. Since it's power requirement will remain low, you will be able to pair it with the still-to-be-seen B boards, so you'll have quite the potent lil' CPU there. Currently, platform cost makes it's a hard recommendation. Intel really needs to put those B boards out. They're not hurting themselves yet, but I'm sure they're not selling as much as they could.

Cheers!

 

[Gon Freecss]

Well, considering that 100Mhz steps in their lineup will not be massive jumps in performance, I'd say the 8400 will continue being the best P/P from the Covfefe Lake lineup until they phase it out or churn out new stuff. Since it's power requirement will remain low, you will be able to pair it with the still-to-be-seen B boards, so you'll have quite the potent lil' CPU there. Currently, platform cost makes it's a hard recommendation. Intel really needs to put those B boards out. They're not hurting themselves yet, but I'm sure they're not selling as much as they could.

Cheers!

Based on Skylake and Kaby Lake, the 8500 will be the best price/performance out of the i5 lineup.

 

[juanrga]

I find it most curious that a week before the 2600 engineering sample was seen on sisoftware we had the leak of an i5-8500@~$199(guess). 3.0GHz base 4.2GHz(6C) and 4.3GHz(1C) boost. Also, with the 2600 it appears to be inline with 10% estimates by GloFo, and looks like Intel is adding segmentations to answer Ryzen products. It will be interesting to see how high they can tweak 2000 series on the top end. 4.4GHz(8C)4.5GHz(1C) for the best case scenario.

The 1800X is 3.6/4.0GHz if the QS and the info about the desktop APU is accurate I think we must expect about 3.8/4.2GHz or 3.9/4.3GHz for the 2800X or whatever was the name. The IMC would increase from 2667 to 2993MHz.

What really intrigues me is James comment that the "400 Series motherboards are designed to improve memory stability and overclocking". This makes me believe that current users of 300-series will be not able to exploit all the potential of the new Pinnacle Ridge CPUs.

 

[Gon Freecss]

What really intrigues me is James comment that the "400 Series motherboards are designed to improve memory stability and overclocking". This makes me believe that current users of 300-series will be not able to exploit all the potential of the new Pinnacle Ridge CPUs.

What about Zen 2 and 3 then? lol

 

[-Fran-]

You should work in marketing, Juan. Why are you saying "unlock all the potential of Pinnacle Ridge CPUs" when talking about the MoBo circuitry helping the IMC to get more stable connectivity for the RAM?

Are you implying the IMC depends, somehow, on the 400-series chipset for some of it's functionality?

 

[juanrga]

You should work in marketing, Juan. Why are you saying "unlock all the potential of Pinnacle Ridge CPUs" when talking about the MoBo circuitry helping the IMC to get more stable connectivity for the RAM?

Are you implying the IMC depends, somehow, on the 400-series chipset for some of it's functionality?

The DIMM modules don't connect directly to the CPU. The signals from the modules to the CPU and viceverse have to travel though the circuitry and wires in the mobo. If the mobo cannot produce fast enough signals or signals aren't stable, then it matter little how many improvements you do to CPU.

The AMD guys is clever: "400 Series motherboards are designed to improve memory stability and overclocking". So, by design, you don't get those improvements in older 300-series mobos.

 

[juanrga]

What really intrigues me is James comment that the "400 Series motherboards are designed to improve memory stability and overclocking". This makes me believe that current users of 300-series will be not able to exploit all the potential of the new Pinnacle Ridge CPUs.

What about Zen 2 and 3 then? lol

I don't know any official statement that claims 300-series mobos will work for Zen2 products, still less Zen3. All what I know is AMD confirming that will stay with the same socket AM4 for Zen2 and Zen3.

 

[-Fran-]

You should work in marketing, Juan. Why are you saying "unlock all the potential of Pinnacle Ridge CPUs" when talking about the MoBo circuitry helping the IMC to get more stable connectivity for the RAM?

Are you implying the IMC depends, somehow, on the 400-series chipset for some of it's functionality?

The DIMM modules don't connect directly to the CPU. The signals from the modules to the CPU and viceverse have to travel though the circuitry and wires in the mobo. If the mobo cannot produce fast enough signals or signals aren't stable, then it matter little how many improvements you do to CPU.

The AMD guys is clever: "400 Series motherboards are designed to improve memory stability and overclocking". So, by design, you don't get those improvements in older 300-series mobos.

Yeah, that is like saying "the new Honda Civic gets better mileage than the previous year version thanks to engine improvements" and being angry about it that they're not making a recall to swap the engine or put those improvements into the new one, or blaming Honda for improving the design and not making it a whole different car.

 

[jdwii]

I can guarantee if the board can handle it in terms of VRM Zen 2 and Zen 3 will come with bios updates to the 300 series boards more specifically x370 with true 6 phase or higher designs.

Even some 790FX boards supported Bulldozer Amd has been historically good at backwards compatibility the chipset itself has very little to do with it.

I would love Intel to do the same make a nice high-end chipset with maybe tougher VRM regulations that company's much follow and let it live for 3-5 years with upgrades.

 

[8350rocks]

"30%-40%" performance means very little at what frequency at what task?

Core for core its more about 10-15% lower IPC with 4-4.2ghz top OCs vs intel which is getting 1ghz more clock speed in a lot of cases when it comes to overclocking.

Also Ryzen is far behind Intel in AVX2 operations

The IPC gap span from 10% in benches more favorable to RyZen muarch to 30% in benches more favorable to CoffeeLake muarch.

The average IPC gap is around 15% in applications. If you multiply this IPC gap by the stock clock gap you obtain a performance deficit of 33.6%. This is the reason why a 8700k is able to match a 1800X in multithreaded applications such as Blender, Handbrake,... despite the Ryzen chip has 33% more cores.

In games the gap can be even higher. I have seen the 8700k beating the 1800X by more than 60% in some DX12 game.

Source? I have not been able to find a single benchmark that shows Ryzen is behind 30% clock-for-clock. Highest I have seen clock-for-clock is 12%.

I am not talking about you taking some theoretical i3 benchmarks and trying to extrapolate some cocktail napkin math IPC gap. I am talking about processor to processor comparisons showing a gap in real world performance.

Hello there, bias.

The 6900K is >45% faster than the slightly higher clocked 1700X in WinRAR. The 7960X is also >50% faster than the 1950X despite having a lower base clock, and slightly slower all core turbo speeds.

Even in the tasks where AMD performs the best, they outperform Intel ever so slightly clock for clock.

Where are the clock speeds? I said clock for clock. If you show me benchmarks with both processors running at 4.0 GHz and the gap is that large, then I will gladly agree with you.

 

[jdwii]

I find it most curious that a week before the 2600 engineering sample was seen on sisoftware we had the leak of an i5-8500@~$199(guess). 3.0GHz base 4.2GHz(6C) and 4.3GHz(1C) boost. Also, with the 2600 it appears to be inline with 10% estimates by GloFo, and looks like Intel is adding segmentations to answer Ryzen products. It will be interesting to see how high they can tweak 2000 series on the top end. 4.4GHz(8C)4.5GHz(1C) for the best case scenario.

So the 8400 will have no point.

Well, it's still a little cheaper and will probably be really close in performance. I think Intel wants to have a chip to compete against every price bracket Ryzen would have something to offer so just to have a counter. They could even drop the 8400 down a few more dollars and really put a hurting on Ryzen. Think of the 8400 around $168 or so... That would put a hurting on Ryzen. Not like they are doing all that well with Intel holding ~90% market.
i5-8400
Recommended Customer Price
$182.00 - $187.00

Until Intel comes out with their mainstream more budget boards i don't even see them in the picture(edit when it comes to budget) unless you are buying a prebuilt machine which sadly in 2018 is getting easier to recommend

 

[Gon Freecss]

"30%-40%" performance means very little at what frequency at what task?

Core for core its more about 10-15% lower IPC with 4-4.2ghz top OCs vs intel which is getting 1ghz more clock speed in a lot of cases when it comes to overclocking.

Also Ryzen is far behind Intel in AVX2 operations

The IPC gap span from 10% in benches more favorable to RyZen muarch to 30% in benches more favorable to CoffeeLake muarch.

The average IPC gap is around 15% in applications. If you multiply this IPC gap by the stock clock gap you obtain a performance deficit of 33.6%. This is the reason why a 8700k is able to match a 1800X in multithreaded applications such as Blender, Handbrake,... despite the Ryzen chip has 33% more cores.

In games the gap can be even higher. I have seen the 8700k beating the 1800X by more than 60% in some DX12 game.

Source? I have not been able to find a single benchmark that shows Ryzen is behind 30% clock-for-clock. Highest I have seen clock-for-clock is 12%.

I am not talking about you taking some theoretical i3 benchmarks and trying to extrapolate some cocktail napkin math IPC gap. I am talking about processor to processor comparisons showing a gap in real world performance.

Hello there, bias.

The 6900K is >45% faster than the slightly higher clocked 1700X in WinRAR. The 7960X is also >50% faster than the 1950X despite having a lower base clock, and slightly slower all core turbo speeds.

Even in the tasks where AMD performs the best, they outperform Intel ever so slightly clock for clock.

Where are the clock speeds? I said clock for clock. If you show me benchmarks with both processors running at 4.0 GHz and the gap is that large, then I will gladly agree with you.

Compared the 6900K to the 1700 and the 1700X. The 6900K is 3.2-3.7GHz, the 1700 is 3.0-3.7GHz, and the 1700X is 3.4-3.8GHz.

 

[juanrga]

Where are the clock speeds? I said clock for clock. If you show me benchmarks with both processors running at 4.0 GHz and the gap is that large, then I will gladly agree with you.

You have a collection of individual benchmarks and averages for both Windows and Linux here

http://www.tomshardware.co.uk/forum/id-3341285/amd-naples-server-cpu-info-rumours/page-25.html#20624519

All them clock-for-clock.

For Windows
Audacity --> 32% IPC gap
Hitman --> 29% IPC gap
7-zip --> 37.5% IPC gap
WinRAR --> 57.1% IPC gap
Adobe --> 32% IPC gap

For Linux
The Stilt got several examples where the IPC gap is higher than 30%. That is the reason why he gives an average of 128.94% for the IPC gap between Zen and Kabylake.

 

[8350rocks]

Where are the clock speeds? I said clock for clock. If you show me benchmarks with both processors running at 4.0 GHz and the gap is that large, then I will gladly agree with you.

You have a collection of individual benchmarks and averages for both Windows and Linux here

http://www.tomshardware.co.uk/forum/id-3341285/amd-naples-server-cpu-info-rumours/page-25.html#20624519

All them clock-for-clock.

For Windows
Audacity --> 32% IPC gap
Hitman --> 29% IPC gap
7-zip --> 37.5% IPC gap
WinRAR --> 57.1% IPC gap
Adobe --> 32% IPC gap

For Linux
The Stilt got several examples where the IPC gap is higher than 30%. That is the reason why he gives an average of 128.94% for the IPC gap between Zen and Kabylake.

I see no benchmarks there...

 

[Gon Freecss]

Where are the clock speeds? I said clock for clock. If you show me benchmarks with both processors running at 4.0 GHz and the gap is that large, then I will gladly agree with you.

You have a collection of individual benchmarks and averages for both Windows and Linux here

http://www.tomshardware.co.uk/forum/id-3341285/amd-naples-server-cpu-info-rumours/page-25.html#20624519

All them clock-for-clock.

For Windows
Audacity --> 32% IPC gap
Hitman --> 29% IPC gap
7-zip --> 37.5% IPC gap
WinRAR --> 57.1% IPC gap
Adobe --> 32% IPC gap

For Linux
The Stilt got several examples where the IPC gap is higher than 30%. That is the reason why he gives an average of 128.94% for the IPC gap between Zen and Kabylake.

I see no benchmarks there...

He linked them above Also, what about the ones I linked?

 

[daerohn]

is it really an IPC gap or an optimization issue? Ofcourse programmers tend to optimize their code according to the market sales. and Intel has been the market leader for nearly a decade, surpassing AMD sales with huge margin. I am not an expert but the cause of this IPC gap can be defined like "Hardware IPC+Software Efficiency". And if we take this into consideration, how much performance can be gained if the code is to be optimized for Ryzen? will it be still the same? or do we have some opportunity here?

 

[juanrga]

is it really an IPC gap or an optimization issue? Ofcourse programmers tend to optimize their code according to the market sales. and Intel has been the market leader for nearly a decade, surpassing AMD sales with huge margin. I am not an expert but the cause of this IPC gap can be defined like "Hardware IPC+Software Efficiency". And if we take this into consideration, how much performance can be gained if the code is to be optimized for Ryzen? will it be still the same? or do we have some opportunity here?

Windows code is usually optimized for ancient CPUs, so it cannot use the newest features in modern Intel CPUs and that is why the average IPC gap with RyZen is lower ~15% in windows reviews.

Linux code is usually more optimized for modern CPUs. Moreover, The Stilt compiled the linux code for the latest architectures (including RyZen) and that is why the average IPC he measured increased to ~30%.

In Windows reviews you can find few benchmarks where the IPC gap is around ~30%. I gave some examples above (Audacity, Hitman, 7-zip, WinRAR, Adobe Lightroom). In Linux reviews you can find many benchmarks where the IPC is around 30% and you can benchmarks where the IPC gap is higher than 200%.