AMD FX-8770 and 9000 - 220W TDP, 5 GHz

[parvadomus]

If they could do it with Trinity (with Richland), they can with Vishera. Altough I don't think they will get something past 4.5Ghz stock, out.

 

[Ajay]

AMD knows how to cool >300W TDP GPU's, so personally I have no concerns about their prowess to develop and implement reliable stock cooling solutions for a 220W TDP CPU.

What I don't believe is that AMD would bother to do something like that given that the marketshare for such a CPU would surely be extremely limited in terms of total units the market would absorb globally on an annual basis.

Intel has now released its second-generation 22nm product. AMD needs to focus on getting their 28nm SKUs to the market in volumes, 32nm is not going to take them where they need to go at this point.

A 220W TDP Steamroller 28nm chip will be head-and-shoulders above a 220W TDP Piledriver 32nm chip, especially because it would come with an added microarchitectural bump to IPC.

^This!

 

[MisterMac]

How will any motherboard run this?

The phases of the normal AM3+ boards sure as hell can't take this?

Specs are crazy - but something will probably be announced @ E3.

 

[Plimogz]

A high OC on say, an 8350 would amount to this sort of TDP, anyway. So I figure that any board known to be capable of running an overclocked+overvolted high-range bulldozer/pilldrier would be a good candidate.

 

[galego]

A high OC on say, an 8350 would amount to this sort of TDP, anyway. So I figure that any board known to be capable of running an overclocked+overvolted high-range bulldozer/pilldrier would be a good candidate.

Some people is already running 8350 @ 5GHz on air. Therefore no problem here with current mobos. Maybe a series of Centurion-certified mobos could be released with those chips .

 

[aigomorla]

When i saw the 220W TDP

Spoiler

*watches people go break out Bong Coolers again*

A high OC on say, an 8350 would amount to this sort of TDP, anyway. So I figure that any board known to be capable of running an overclocked+overvolted high-range bulldozer/pilldrier would be a good candidate.

Thats the difference tho.. a HIGH OC.... and a lot of fine tuning... u probably wont hit 220W TDP on a 100W TDP chip also.

220W TDP means this chip is guzzling 220W on STOCK.

 

[inf64]

I've just read sweclocker's review of 4770K/4670K. I wanted to see what kind of difference, power wise, are we looking at Vs 8350. It turns out that in OCCT 4770K draws 140W vs 210W for 8350, both at stock. When OCed to 4.4Ghz, the power draw of 4770K spikes to 190W while performance increases roughly by ~11% or so (logical since the chip almost always tries to run at full turbo of 3.9Ghz).

Now when we look at the rumored OCed-out-of-the-box FX models, we see them having the rather high TDP of 220W. This is 95W higher than official TDP spec for 8350 (or a bit less than that considering 8350 can actually go above 125W). Now looking at sweclocker's link I posted above, the FX9000 @ 5Ghz should be sitting at 210+95~=305W. This is with 20% clock (performance) increase since both rumored stock and Turbo clock are ~20% higher than on 8350. 305W of "stock" FX9000 vs 190W of OCed to rather expected/normal 4.4Ghz for 4770K on air- that's 115W of difference, roughly of course. Since the jump with FX9000 is 20% vs the perf. increase of 10% the OC of 4700K nets, we have the FX9000 actually closing the gap when compared to OCed 4770K. Versus stock 4770K, new "stock" FX should be winning some of the benchmarks and closing the gap significantly in others. In some situations the gap will still be there though and with a hefty power difference at that.

Finally the pricing structure. If AMD prices these parts at lower range than 4770K, somewhere close to 4670K or a bit higher (250-260$), it could be an "OK" purchase for those wanting great MTed performance and somewhat competitive ST performance (5Ghz Vishera is pretty fast chip in ST tasks, still slower than IB and co though). Lastly those buyers should be ready to accept the fact their chip will be drawing >120W more than OCed IB/Haswell setup,but will be cheaper ~100$ or so.

 

[guskline]

Some people is already running 8350 @ 5GHz on air. Therefore no problem here with current mobos. Maybe a series of Centurion-certified mobos could be released with those chips .

If you use the high end Noctua perhaps the 8350 could boot into Windows at 5ghz BUT you need a big voltage increase AND will it be stable for AMD overdrive, Prime95 and IntelBurn test?

The best I could do with my 8350 ( yes galego I actually have a 8350 AND a 8320 chip and despite my requests you haven't even told us what you have) is 4.6 Ghz.

I actually cranked my vcore to 1.52 and booted into windows 8 but the load temps were too high, even with a Corsair H100 water cooling system.

Thus posts, such as "some people is ( should be are) already running 8350 @ 5Ghz on air is suspect at best. There are posters on other web sites who have cranked their 8350s to near 5Ghz or slightly above, but almost everyone runs a very high end water cooling system. Most air coolers cannot handle the termal output when the 8350 is OC'd so high.

 

[Rvenger]

It has been proven that the 6 Core FX does very well performance wise at higher clocks. They should have released a 5ghz 6 core part instead and it would get that TDP in a reasonable range.

 

[Chiropteran]

I'd probably buy one just for the novelty of it, and as a joke for the silly hypocritical environmentalists who cry about wasting a couple watts of electricity while owning a cat or dog that causes far more environmental impact.

 

[os2wiz]

Some people is already running 8350 @ 5GHz on air. Therefore no problem here with current mobos. Maybe a series of Centurion-certified mobos could be released with those chips .

That is completely false. NO BODY has a stable 8350 5.0 GHZX overclock on air. ALL are using liquid coolers to obtain 5.0 GHZ stable. I repeat the word stable. You may boot at 5.0 GHZ in exceptional circumstances on air. You would NEVER be able to successfully complete any serious benchmarks or stress tests on air at that frequency. You are doing a disservice to spread such false information. I am registered on overclock.net and techpowerup.comboth in the Vishera Owners club. I know from both personal experience and reading thousands of posts that you are not being truthful.

 

[Ajay]

I'd probably buy one just for the novelty of it, and as a joke for the silly hypocritical environmentalists who cry about wasting a couple watts of electricity while owning a cat or dog that causes far more environmental impact.

LOL! Thanks for the laugh :thumbsup:

 

[inf64]

Ok guys, finally some computex news (still not official by AMD though). Thanks goes to Hexus' and overclocknet's CPU forums .
It's semi-official, there will be a "high-TDP" FX models presented soon, gigabyte lists it as 5Ghz AMD FX lol
Original source is donanimhaber.

Refer to my previous post above for TDP and performance numbers/comparisons

 

[AtenRa]

Supports AMD next generation of AMD+ 5GHz FX series

Vishera 2.0 ??

 

[sequoia464]

Ok guys, finally some computex news (still not official by AMD though). Thanks goes to Hexus' and overclocknet's CPU forums .
It's semi-official, there will be a "high-TDP" FX models presented soon, gigabyte lists it as 5Ghz AMD FX lol
Original source is donanimhaber.

Refer to my previous post above for TDP and performance numbers/comparisons

The spec for 8 Sata3 ports, suppose those might be native with the long rumored 1090 chipset?

 

[inf64]

Well I doubt it's "next generation" in a sense that uarchitecture changed. They might have some minor success in optimizing the design for even higher clockspeeds (albeit with considerable power draw increase).
I heard some people mentioning the RCM and it's supposed implementation in Richland version of APU and this "vishera 2.0" , but that's just speculation. I always thought that RCM has major impact at much lower clocks and not these corner case situations where power/performance ratio is completely skewed.

One thing is for sure: 4.8/5Ghz "stock" Vishera even with that big TDP rating should be faster than 4770K in many workloads. ST is still not going to be AMD's domain but it will bridge the gap somewhat. Brute force 20% performance increase FTW LoL . I wonder whether they will bundle some serious WC kits with these things, Performance might be there but cooling this thing (and VRMs on these boards!) is going to be a challenge I bet!

 

[Ajay]

Well I doubt it's "next generation" in a sense that uarchitecture changed. They might have some minor success in optimizing the design for even higher clockspeeds (albeit with considerable power draw increase).
I heard some people mentioning the RCM and it's supposed implementation in Richland version of APU and this "vishera 2.0" , but that's just speculation. I always thought that RCM has major impact at much lower clocks and not these corner case situations where power/performance ratio is completely skewed.

One thing is for sure: 4.8/5Ghz "stock" Vishera even with that big TDP rating should be faster than 4770K in many workloads. ST is still not going to be AMD's domain but it will bridge the gap somewhat. Brute force 20% performance increase FTW LoL . I wonder whether they will bundle some serious WC kits with these things, Performance might be there but cooling this thing (and VRMs on these boards!) is going to be a challenge I bet!

I just have to believe it's a new stepping (like the famous D0 Ci7 920s), otherwise is seems crazy. That, or AMD has been binning these things for a while in anticipation of Haswell, even then, I can't see this CPU running at anywhere near 220W except under light usage.

Kudos for the effort though. This is the sort of thing I would expect from Jerry Sanders. Maybe in getting lean AMD has gotten pretty feisty too!

 

[Sweepr]

Theres no denying they're trying. Intel should follow the same path and release 150W+ TDP 8C-10C IB-E instead of another rounds of hexa-cores for enthusiasts.
I wonder if there's still some overclocking headroom left on air for that 220W Vishera. Given that i5-4670K is surprisingly close to FX8350 even in highly MT apps (~2600K level of performance without HT because of IPC bump) a +4.2GHz i5-4670K would beat this 220W TDP monster in many tasks, probably using a lot less power too.

http://www.hardware.fr/medias/photos_news/00/41/IMG0041515.png

 

[inf64]

You have to take into consideration that Haswell almost never runs at base clock but tries to use as much of the max. Turbo clock as possible. So 4.2Ghz clock 4C/4T Haswell wouldn't be 4.2/3.4=1.23 or 23% faster but something closer to half of that.
According to the chart from hardware.fr , 4670K scored 164.9 while 8350 scored 171pts. Let's be generous and give 4.2Ghz +15% above "stock": 164.9x1.15=190pts.
5Ghz FX would score ~19.5% above stock 8350: 171x1.195=204.4pts. So no, 4.2Ghz 4C/4T Haswell wouldn't be able to touch 5Ghz FX. Power wise the difference would probably be around 120W or so, so less performance but less power too.

 

[SiliconWars]

It's clear that globalfoundries process keeps improving a lot and this could be one of the main factors. If you remember back when BD was released the process seemed to have a hard ceiling of 4.3-4.4 GHz GHz turbo that didn't really make sense if you looked at what the CPU's were capable of on 8 cores.

 

[parvadomus]

Ok guys, finally some computex news (still not official by AMD though). Thanks goes to Hexus' and overclocknet's CPU forums .
It's semi-official, there will be a "high-TDP" FX models presented soon, gigabyte lists it as 5Ghz AMD FX lol
Original source is donanimhaber.

Refer to my previous post above for TDP and performance numbers/comparisons

LOL I have to see this to believe. Its even more interesting than the Failwell launch. D:

 

[Abwx]

Historicaly they extracted 30% higher frequency through
a node maturation so they can well reach 5.2Ghz but
they ll have to increase the TDP rating to an extent
that can hardly be estimated with precision , the 220W
figure seems to me quite high , not technicaly but in
respect of sales expectations , i wonder how much
such chips they could sell and the quantity needed
to reach the break even point given the inherent
binning costs.

 

[Sweepr]

You have to take into consideration that Haswell almost never runs at base clock but tries to use as much of the max. Turbo clock as possible. So 4.2Ghz clock 4C/4T Haswell wouldn't be 4.2/3.4=1.23 or 23% faster but something closer to half of that.
According to the chart from hardware.fr , 4670K scored 164.9 while 8350 scored 171pts. Let's be generous and give 4.2Ghz +15% above "stock": 164.9x1.15=190pts.
5Ghz FX would score ~19.5% above stock 8350: 171x1.195=204.4pts. So no, 4.2Ghz 4C/4T Haswell wouldn't be able to touch 5Ghz FX. Power wise the difference would probably be around 120W or so, so less performance but less power too.

So 7,5% faster in a series of highly MT taks consuming 120W more power? Seems like a good trade-off (and many 4C/4T Haswell i5 should be able to reach ~4.4GHz so I'm being generous here). Of course theres always 4C/8T Haswell when you need it (around 200 pts in Hardware.fr chart @ stock), but I'm guessing AMD will price these chips lower than them.

Gaming performance will finally come close to a (2011) 2500K @ stock clocks according to the gaming charts (if we assume perfect scaling).

 

[Torn Mind]

PSU and cooling product companies are jumping with joy with this news.

 

[Ajay]

PSU and cooling product companies are jumping with joy with this news.

LOL!