Ryzen 1800x at 5.2Ghz. Sets a new record on Cinebench

[Batmeat]

All I can say is wow!! I'm super excited for my new build.

link

 

[jihe]

What world record? Xeon 2699v4 at stock gets well over 2600 in cinebench.

 

[IEC]

Record specifically for 8-core processors. Beating the 6900K and 5960X.

 

[UsandThem]

Before the average hardware enthusiast gets their hopes up too much.

The screenshot of the test shows the CPU speed reaching 5201.07 MHz, thanks to a core voltage of 1.875V, the bus speed at 137.78 MHz and the multiplier up at 37.75x. Of course, this was done by professionals, and with the help of some liquid nitrogen cooling taking the temperature down to -200C

 

[ClockHound]

Before the average hardware enthusiast gets their hopes up too much.

Wait. What? Are you implying that Ryzen won't be shipping with its own stock PWM-controlled LN2 cooling solution? Oh..man...what a let down.

;-)

 

[riggnix]

Is it just me, or does 5.2 on nitrogen sound kinda disappointing? For water (or even air) it would be pretty impressive, but on nitrogen this sounds a little bit like a hard cap... Can't wait to see OC reviews...

 

[LTC8K6]

Given the 4.1/4.2 results on water, it sounds about right?

 

[riggnix]

I don't think, that 4.2 is even close to max on water. Stock boost clock is 4.0. 200 MHz OC from stock cooler to water? I don't think so. At least I don't hope so, that would be really bad.

 

[lolfail9001]

I don't think, that 4.2 is even close to max on water. Stock boost clock is 4.0. 200 MHz OC from stock cooler to water? I don't think so. At least I don't hope so, that would be really bad.

When your justification is "would be really bad", you are in for a bad day.

 

[riggnix]

When your justification is "would be really bad", you are in for a bad day.

Well, let's see. When thinking about all we currently know about the architecture I don't see why there should be such a low frequency cap. But of course I might be missing something.

I would also really be interested in the mainboards used for the leaked OCs.

Let's just say I can't wait for "real" reviews of the final products to finally know for sure. Let's just hope it won't be another Phenom I, clockwise.

 

[french toast]

We know scan is selling pre overclocked systems @4.2 on water? Make your own mind up whether thats max, i know what i think.
In the advent that it IS 4.2 max, we know R7 1700 is capable of 4050mhz on water and crosshair hero, so what a bargain that 329$ cpu is

 

[riggnix]

It seems to me, that all of the "bad" oc results that are being leaked are not 1800X but other models. This could also indicate heavy binning of chips. This would make a lot of sense, because it's a pretty new process for GF... If these really are max OCs, than I'm gonna get a 1700 for sure, that one is just too good to be true then.
But maybe there is such a big premium in the 1800X because it's cherry picked chips that clock considerably higher.

 

[french toast]

It seems to me, that all of the "bad" oc results that are being leaked are not 1800X but other models. This could also indicate heavy binning of chips. This would make a lot of sense, because it's a pretty new process for GF... If these really are max OCs, than I'm gonna get a 1700 for sure, that one is just too good to be true then.
But maybe there is such a big premium in the 1800X because it's cherry picked chips that clock considerably higher.

Yea, later on in the year i might go R7 1700, crosshair hero and blow ram budget og one stick of 16mb high end ddr4 3600-4200mhz memory (yes single chanel)

Then in two years i can grab 7nm zen++ and another stick of ram which will be Much cheaper.
Edit if there are 4 slots i could go 2x 8gb dimms and have dual channel.

 

[riggnix]

Yea, later on in the year i might go R7 1700, crosshair hero and blow ram budget og one stick of 16mb high end ddr4 3600-4200mhz memory (yes single chanel)

Then in two years i can grab 7nm zen++ and another stick of ram which will be Much cheaper.
Edit if there are 4 slots i could go 2x 8gb dimms and have dual channel.

Sounds reasonable. There has to be A LOT more OC potential for me to go for the R7 1800X. Also Gigabyte caught my eye with the GA-X370 Gaming 5 (http://www.gigabyte.us/Motherboard/GA-AX370-GAMING-5-rev-10). Seems to have the same specs as the higher-end K7, only with different coloring. Much cheaper than the Crosshair, but seems like a very decent board.

 

[french toast]

Sounds reasonable. There has to be A LOT more OC potential for me to go for the R7 1800X. Also Gigabyte caught my eye with the GA-X370 Gaming 5 (http://www.gigabyte.us/Motherboard/GA-AX370-GAMING-5-rev-10). Seems to have the same specs as the higher-end K7, only with different coloring. Much cheaper than the Crosshair, but seems like a very decent board.

yea good choice, hopefully by the time ive saved some money the crosshair will be reduced, that way i get two generations of cpu with overclocking with a rock solid mobo, potentially 8 years.

 

[LTC8K6]

I don't think, that 4.2 is even close to max on water. Stock boost clock is 4.0. 200 MHz OC from stock cooler to water? I don't think so. At least I don't hope so, that would be really bad.

Just judging by AMD's recent demonstration of XFR and manual overclocking of an 1800X on water. XFR gave them 100mhz to 3.7 all core and 4.1 single core, and 4.1 is what they tried to set all cores to manually. We don't know what went wrong, but the 4.1 manual overclock was not shown, as the feed/computer crashed.

So if AMD got 100mhz, then 200mhz is probably a reasonable assumption for being near the limit with a good water cooling system.

The other possibility is that AMD sandbagged the demo, but I wonder if that is really likely?

 

[riggnix]

Just judging by AMD's recent demonstration of XFR and manual overclocking of an 1800X on water. XFR gave them 100mhz to 3.7 all core and 4.1 single core, and 4.1 is what they tried to set all cores to manually. We don't know what went wrong, but the 4.1 manual overclock was not shown, as the feed/computer crashed.

So if AMD got 100mhz, then 200mhz is probably a reasonable assumption for being near the limit with a good water cooling system.

The other possibility is that AMD sandbagged the demo, but I wonder if that is really likely?

Might have been unstable mobo firmware. It was on the crosshair, right? If the stability of early versions for am2+ and am3 Mainboards by Asus are any indication, that might get much better in the next few versions.

But that's starting to sound like wishful thinking even to me

 

[Shamrock]

Sounds reasonable. There has to be A LOT more OC potential for me to go for the R7 1800X. Also Gigabyte caught my eye with the GA-X370 Gaming 5 (http://www.gigabyte.us/Motherboard/GA-AX370-GAMING-5-rev-10). Seems to have the same specs as the higher-end K7, only with different coloring. Much cheaper than the Crosshair, but seems like a very decent board.

There is something interesting about this board.
https://www.newegg.com/Product/Product.aspx?Item=N82E16813128992

Look ALL the way down at the bottom of the specs. Gigabyte will "officially" support Win 7? It's the only board that have said this AFAIK.

Operating System
- Support for Windows 10 64-bit
- Support for Windows 7 64-bit
* Please download the "Windows USB Installation Tool" from GIGABYTE's website and install it before installing Windows 7.

 

[unseenmorbidity]

Just judging by AMD's recent demonstration of XFR and manual overclocking of an 1800X on water. XFR gave them 100mhz to 3.7 all core and 4.1 single core, and 4.1 is what they tried to set all cores to manually. We don't know what went wrong, but the 4.1 manual overclock was not shown, as the feed/computer crashed.

So if AMD got 100mhz, then 200mhz is probably a reasonable assumption for being near the limit with a good water cooling system.

The other possibility is that AMD sandbagged the demo, but I wonder if that is really likely?

xfr all core boost is likely capped at 100mhz

I suspect a range of about 3.9-4.2 GHz for 1700, 1700x, and 1800x. Assuming a good board and cooling solution.

 

[KTE]

This confirms what I said all along. It's a high IPC chip on a LPP process shipping right at its limit.

5GHz is 2GHz less than SkyLake FMax with the same cooling, just for perspective. Almost like K8 vs P4 here.

Sent from HTC 10
(Opinions are own)

 

[lolfail9001]

5GHz is 2GHz less than SkyLake FMax with the same cooling, just for perspective.

Actually the FMax on LN2 looks to be 1.5Ghz lower than Kaby Lake's. I mean, that's a ton, without a question, but just correcting the perspective here.

 

[Zucker2k]

This confirms what I said all along. It's a high IPC chip on a LPP process shipping right at its limit.

5GHz is 2GHz less than SkyLake FMax with the same cooling, just for perspective. Almost like K8 vs P4 here.

Sent from HTC 10
(Opinions are own)

I was one of those who was particularly interested in your back and forth with bjt2 and his rather over-optimistic, and sometimes, wild frequency claims (4.7Ghz, anyone?). I wonder what he has to say about that now. I'd still wait for final reviews though. But to think even with XFR in place, they could only gain 100mhz is really interesting to say the least. Interesting 24 hours ahead.

 

[vissarix]

Meanwhile the overhyped R7 1700x loses to the i7 6800k on gaming benchmarks
http://wccftech.com/amd-ryzen-7-1700x-1700-official-gaming-benchmarks-leak/

another hype train derailed..

Trolling is not allowed
Markfw
Anandtech Moderator

 

[MongGrel]

Given the 4.1/4.2 results on water, it sounds about right?

I do not know what you are inferring by it sounds right.

Win7 x64 support, they are really stepping up the game there

 

[LTC8K6]

I do not know what you are inferring by it sounds right.

Win7 x64 support, they are really stepping up the game there

I'm not inferring anything. Given the AMD 1800X demo, the numbers sound right to me.

I thought the Win7 support idea for Zen was quashed?