AMD Ryzen SKU and Price Information/Speculation.

[dacostafilipe]

Custom water cooling the Ryzen 1800x sounds like a good idea ...

I'm still deciding if I should go with an AIO or a custom loop ... as I'm kinda getting lazy over the years, it will certainly be AIO

 

[Lovec1990]

I'm still deciding if I should go with an AIO or a custom loop ... as I'm kinda getting lazy over the years, it will certainly be AIO

AIO is not that good air cooling is better so its air or custom

 

[Dresdenboy]

The cooler can handle way more than 125W. Its 150W capable at least and I would think 175W too. The reason why is in the extreme low Tcase rating for the X chips. I cant remember seeing one below 60C Tcase before. You are right tho, they would be 125W TDP chips without very good cooling. Will be interesting to see how 3rd party coolers that are used to much higher Tcase ratings will handle it.

Its all related to this:

Thanks for the graphs. They support my point, that lower temps give more room for XFR. What you might have gotten wrong are cause and effect. You already deduce that the X models would need that better cooler to stay inside their 95W target, while contrary to that there is the possibility, that XFR will simply perform better with better cooling provided. X models without that cooler (WOF) will leave the choice for the appropriate cooling solution to the customer.

Here's an interview, where Hallock stated, that the environment temps needed to get the stated base/boost clocks are 32°C (90°F) - likely one point, where XFR will begin to kick in (aside from other factors as usual). Anything below that would result in higher XFR clocks.

Robert Hallock: ... will be able to do that, no question. To return to the Extended Frequency Range: We set the ambient temperature for Base and Boost Clock to 32 degrees Celsius or around 90 degrees Fahrenheit. At home, I have 70 and in my PC 78 degrees Fahrenheit - so I have thermal headroom from the start.

http://www.pcgameshardware.de/AMD-Zen-Codename-261795/Specials/Interview-Ryzen-1218658/

More on that topic:

Even then, there may be performance boosts on offer for those with exotic cooling as part of its "Extended Frequency Range" (XFR) feature.

"A lot of processors have pre-programmed clock speed voltage tables," explained AMD's Robert Hallock at CES. "We don't. This is very algorithmic. We analyse power consumption limits, thermal limits, silicon utilisation limits, and out of that boundary, if none of those limits are being met, you can just keep raising clock speed until one of them is. Then you level off the boost and then try to sustain it as long as possible. The system is smart enough to know what's going on inside itself, and adaptive enough to prevent sudden drops in clock speeds."

https://arstechnica.com/gadgets/2017/01/amd-ryzen-motherboards-hype/

 

[majord]

It will require all of that 120mm AIO when I get my dirty hands on it

I'm surprised they're not launching the XFR models with an AIO SKU to be honest. ala FX 9370/50 . That was a great kit, would be particuarly good with the CPU horsepower to match.

 

[RichUK]

AIO is not that good air cooling is better so its air or custom

Air cooling is better than an AIO?

Mmmmkay

 

[lobz]

AIO is not that good air cooling is better so its air or custom

All my air and water cooling solutions I've ever owned beg to differ.

 

[realibrad]

All my air and water cooling solutions I've ever owned beg to differ.

Most AIOs (assuming he meant closed loop) cannot beat a Noctua NH-D15 and if they do its a small amount. The biggest issue to me is that a D14 and D15 weigh a ton for a motherboard whereas a closed loop does not put the same stress on the board.

http://www.relaxedtech.com/reviews/noctua/nh-d15-versus-closed-loop-liquid-coolers/2

Closed loops are getting better and cheaper, so I would expect that to change over time.

 

[Agent-47]

AIO is not that good air cooling is better so its air or custom

Could not disagree more.

 

[Crumpet]

Air cooling > 120mm AIO
240mm+ AIO > Air Cooling

 

[Agent-47]

Most AIOs (assuming he meant closed loop) cannot beat a Noctua NH-D15 and if they do its a small amount.
http://www.relaxedtech.com/reviews/noctua/nh-d15-versus-closed-loop-liquid-coolers/2

.

The link you posted shows otherwise by 3-4C at load. Albeit at a higher cost.

 

[realibrad]

The link you posted shows otherwise by 3-4C at load. Albeit at a higher cost.

My link does not show otherwise. 2 of the 5 top closed loops could beat the Noctua. Those are not the only closed loop coolers, and are at the upper end of those systems. Once you realize that the list is the top of the top, and then see the Noctua right there, then you see that my point was backed up. Also keep in mind that the Noctua can go further than the tests, because the Noctua used quieter fans. The only ones that beat it were way louder. As fan speed increases air flow increases and noise increases. If the Noctua cooler had higher airflow fans it likely would have beaten those systems.

 

[Agent-47]

My link does not show otherwise. 2 of the 5 top closed loops could beat the Noctua. Those are not the only closed loop coolers, and are at the upper end of those systems. Once you realize that the list is the top of the top, and then see the Noctua right there, then you see that my point was backed up. Also keep in mind that the Noctua can go further than the tests, because the Noctua used quieter fans. The only ones that beat it were way louder. As fan speed increases air flow increases and noise increases. If the Noctua cooler had higher airflow fans it likely would have beaten those systems.

Err. Two AIO cooler better than d15, arguably the best air cooler. One tie. Two below. That does not equat to " Most AIOs cannot beat a Noctua NH-D15 " in my book. There are lots of arguments against AIO, but performance is not one of them.

If sound is an issue please get a couple of noctua fans, that's what I do.

 

[Agent-47]

http://wccftech.com/amd-ryzen-full-lineup-pricing-clock-speeds-leaked/

Is this for real?

No quad core above 3.8/9 GHz. Seems like we have a slight clocking issue?

Other interesting thing is 8c16t 3.7 GHz is 65w, but a 6c12t above 3.5 is 95w? Is boast clock above TDP? I thought that was a XFR thing.

 

[realibrad]

Err. Two AIO cooler better than d15, arguably the best air cooler. One tie. Two below. That does not equat to " Most AIOs cannot beat a Noctua NH-D15 " in my book. There are lots of arguments against AIO, but performance is not one of them.

If sound is an issue please get a couple of noctua fans, that's what I do.

I am mistaken. Those 5 AIOs are apparently the only AIOs on the market. The tested AIOs do not represent the best of the best AIOs and as such fully disprove that the Noctua D15 is typically better than AIOs. I have made an error in my judgement in believing that there were more than 5 AIOs in the market.

 

[Erenhardt]

http://wccftech.com/amd-ryzen-full-lineup-pricing-clock-speeds-leaked/

Is this for real?

That XFR is quite expensive if true

Also, R5 1300 seems like a steal.

 

[Glo.]

That XFR is quite expensive if true

Also, R5 1300 seems like a steal.

They made mistake. 6 core 65W is 3.3/3.7 GHz. Its validation silicon. 95W, will only clock higher. And its most likely, the CPU that has leaked on Taobao, or whatever that Asian supplier is called. And it has shown core clock of 4.2 GHz.

 

[StinkyPinky]

These latest leaks seem too good to be true to be honest. Like Agent Mulder, I want to believe. I really do. But there's this little voice in my brain that whispers over and over again that AMD products have been hyped before and failed to deliver

 

[Glo.]

This is how clueless sometimes is WTFTech:
Article on pricing yes? http://wccftech.com/amd-ryzen-full-lineup-pricing-clock-speeds-leaked/
Lets look at something:

R5 1600X, is here shown as 65W TDP chip, for 1999Y. Everything is fine? Nope.
First:

1600X is 95W+ TDP Chip.
Secondly, if the price is correct, 1999Y, then:

It has 4.2 GHz boost clock. Which should be logical, considering, that 1800X is 95W+ TDP, and has 4.0 GHz.

And lastly. CanardPC, have said, that currently AMD was running 6C/12T at 3.3GHz base clock@65W TDP. So something, somewhere is off. Maybe just an error of communication. However, be sure, that 95W TDP 6C/12T CPU is not just 3.3/3.7 GHz CPU.

And to end this: there has never leaked any 6C Eng sample/qualification Sample that have had lower than 3.3 GHz base clock.

 

[Agent-47]

These latest leaks seem too good to be true to be honest. Like Agent Mulder, I want to believe. I really do. But there's this little voice in my brain that whispers over and over again that AMD products have been hyped before and failed to deliver

Granted they are priced a lot more favourably, but at those clocks I don't think a 4c4t/4c4t amd CPU can out do a Intel on gaming, specially the once that are not properly threaded. Other than that I think it will be a killer line of CPUs.

 

[SK10H]

http://wccftech.com/amd-ryzen-full-lineup-pricing-clock-speeds-leaked/

Is this for real?

No quad core above 3.8/9 GHz. Seems like we have a slight clocking issue?

Other interesting thing is 8c16t 3.7 GHz is 65w, but a 6c12t above 3.5 is 95w? Is boast clock above TDP? I thought that was a XFR thing.

The price of the SKUs looks quite familiar to me, 8C16T at i7 price, 6C12T at i5 price, like I told *you* so to the usual people last April.
http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=threads/new-zen-microarchitecture-details.2465645/page-24#post-38141449

 

[dawgmatix]

Is there any information available for server chips based on Zen (Naples). Will Zen be available in four way configurations - ie with motherboards which take 4 Zen based server chips?

 

[Face2Face]

Can't wait to throw some more cores at my Plex sever. I'll probably go with the lowest wattage 6/12 option.

 

[Timmah!]

I guess i grew cynical, seeing prices of hardware grow steadily year by year, so i have hard time believing now these rumors...or better said, its incredibly suspicious to me, that AMD would sell their new 8-core CPUs which are rumored to be pretty much equal in performance to Intel´s 8-core CPUs almost for the price of Intel 4-core CPUs... something does not compute. Either they are not going to be as fast as suggested, or these prices gotta be wrong. I did not expect AMD to be expensive as Intel obviously, but selling comparable product for 1/3 of its price? Why would they do that?

 

[Glo.]

I guess i grew cynical, seeing prices of hardware grow steadily year by year, so i have hard time believing now these rumors...or better said, its incredibly suspicious to me, that AMD would sell their new 8-core CPUs which are rumored to be pretty much equal in performance to Intel´s 8-core CPUs almost for the price of Intel 4-core CPUs... something does not compute. Either they are not going to be as fast as suggested, or these prices gotta be wrong. I did not expect AMD to be expensive as Intel obviously, but selling comparable product for 1/3 of its price? Why would they do that?

Marketing. They want to build their brand. When Kia was releasing very good, modern car, at that time, the Cee'd, it was great value. One of best cars in the class, and was priced much better than its counterparts. AMD is doing the same thing, with their CPUs.

And they are going to have enough margin on those CPUs.

 

[AtenRa]

They do require a "Wraith" class cooler to stay within specs. Its not going to work with a HS65 95W cooler for example.

They dont require a Wraith cooler for normal operation, they can use the HS65 95W cooler.