AMD K10 SuperPi Reults

[T2k]

http://www.xtremesystems.org/forums/...&postcount=100

Wow, 39 vs 41 secs - that'd be REALLY-REALLY crap for a new architecture and for AMD stocks...

[informal]

Wow, 39 vs 41 secs - that'd be REALLY-REALLY crap for a new architecture and for AMD stocks...

Relax,the results are not final:

http://www.xtremesystems.org/forums/...&postcount=155

[uOpt]

Thanks.

Now we only need a 64 bit Cinebench score for K8.

SuperPi really isn't too meaningful for CPU performance.

[T2k]

Relax,the results are not final:

http://www.xtremesystems.org/forums/...&postcount=155

Well that changes the whole picture - thanks for the link; I'll put it into my sigline.

[AliG]

Relax,the results are not final:

http://www.xtremesystems.org/forums/...&postcount=155

good, if we're lucky we'll be seeing about twice the performance of that if amd's claim of 90% over k8 is correct

[uOpt]

Well that changes the whole picture - thanks for the link; I'll put it into my sigline.

I don't see how the chipset and BIOS would have that much impact on a platform with memory controller on the CPU die. It's not that messing with anything but the primary timings, RAM frequency and CPU clock does much to real-world performance for K8.

That's single-core. Of course a lot more happens for the across-core scores.

[Jass]

at 2Ghz... this CPU is useless to me. If my dual core @ 3.2Ghz can hit 25s 1M that that Quad Core is weak. All I want to see is how great it overclocks, and if it goes beyond all other Intel Quad cores or at least gets a similar clock, that would be pretty amazing.

[BeardyMan]

at 2Ghz... this CPU is useless to me. If my dual core @ 3.2Ghz can hit 25s 1M that that Quad Core is weak. All I want to see is how great it overclocks, and if it goes beyond all other Intel Quad cores or at least gets a similar clock, that would be pretty amazing.

there are no intel quad cores, only glued dual cores

[AliG]

at 2Ghz... this CPU is useless to me. If my dual core @ 3.2Ghz can hit 25s 1M that that Quad Core is weak. All I want to see is how great it overclocks, and if it goes beyond all other Intel Quad cores or at least gets a similar clock, that would be pretty amazing.

I think you have it wrong, last time I checked super pi was only single or dual threaded, that's why you see people with x6800s getting better scores than the xq6800 (for clockspeed and because the bus isn't fast enough for the qx6800, while the qx6850 on the other hand is a different story) a lot of the time. As for overclocking, I would be greatful with a 3.2ghz quad, I don't expect amd to hit 4ghz (actually it's been done before, but very rare like a 7.5+ghz p4) until 45nm with high K and metal gates, that should help a bit along with the better soi process the clocks should go up slightly. But until then, be happy it's even being launched, as the actual phenom cpus won't be out for a while

[Trunks]

there are no intel quad cores, only glued dual cores

²that

[T2k]

I don't see how the chipset and BIOS would have that much impact on a platform with memory controller on the CPU die. It's not that messing with anything but the primary timings, RAM frequency and CPU clock does much to real-world performance for K8.

That's single-core. Of course a lot more happens for the across-core scores.

He says chipset, BIOS have changed plus HT activated as well as L2 cache is enabled now on the new chips.

All together it can easily redraw the lines - sound much closer to Intel, to tell you the truth.

[[XC] flat-four]

Wow, 39 vs 41 secs - that'd be REALLY-REALLY crap for a new architecture and for AMD stocks...

I seriously doubt that AMD would spend all this time developing a new processor and would gain this little. From a stricly logical viewpoint, this doesn't make sense. Also, this is Coolaler's site, but who are these people posting the screenshots? At least with THe Inq's article, we at least have someone to hold accountable.

[road-runner]

lol somebody fell for it already, I just stuck it there a couple minutes ago

I have to admit tho I almost smacked my screen not long after putting it there hehe

I almost hit my screen also, thats a good one there...

[metro.cl]

I know that RD790 doesnt have much problems i have all the Errata files, most problems come from HT3 and power features, but all are almost solved

[T2k]

I seriously doubt that AMD would spend all this time developing a new processor and would gain this little. From a stricly logical viewpoint, this doesn't make sense. Also, this is Coolaler's site, but who are these people posting the screenshots? At least with THe Inq's article, we at least have someone to hold accountable.

Gary added some stuff:

1. Around 2.4GHz and higher you will want to run CAS4 1066 and at 3GHz+, we expect/estimate that 1333 CAS5 will come in handy. AMD is working very closely with the memory suppliers at this time to get low latency DDR2-1066 ready quickly and to start looking at DDR2-1333 next year before they worry about the switch to DDR3. Memory latencies are going to be a key with this CPU and the performance oriented consumer chipsets.

2. The current AM2+ boards are still immature from a driver/chipset viewpoint, at least to the point of not providing benchmarks yet, once they get closer, expect some numbers.

3. The lower speed Barcelonas on the server chipsets are not going to shine that well in a lot of consumer applications (against higher clocked Yorkfields, but that is not the target market right now), so AMD desperately needs to get the speeds up for this chip to show its true potential. Right now, its doing a lot better than what we saw at Computex and we understand the latest silicon is a marked improvement (several of the board guys were extremely pleased with the last samples) over the last spin we tested. The numbers will be out in a couple of weeks, some will be very happy, some will not, but at least the damn thing will finally be shipping.

p.s. Not trying to be vague, just until the final CPU samples are in and the green light is given by the board guys, no real point in guesstimating.

[xVeinx]

As was pointed out in an another thread, the HT link versus FSB in the CPU-Z is kinda odd. I was under the impression that this info would be drawn from the chipset/bios, and I'm not sure how that would get screwed up by CPU-Z...

[uOpt]

He says chipset, BIOS have changed plus HT activated as well as L2 cache is enabled now on the new chips.

All together it can easily redraw the lines - sound much closer to Intel, to tell you the truth.

Rubbish. There is no way the L2 cache was off in these tests. Hypertransport? OK, but it won't play a role for superpi and very little role for the per-core score in Cinebench.

[uOpt]

I seriously doubt that AMD would spend all this time developing a new processor and would gain this little.

I and several other people have reviewed the list of actual changes to the K10 integer units and I came up with a prediction of 5% win over K8 per-core same-clock. I have documented this prediction here and at 2cpu and you won't see my posts haven't been edited since then.

The new L3 cache won't do much for per-core performance since L2 is already big enough for most real-world applications common demands (it was in K8) and going through an additional level of cache has a cost associated to it.

The point about K10 is to delivery many-core low-power systems with good IPC in the server segment and quad-core low-cost systems in the desktop and home use segment.

The huge and unexpected win that Core2 had over K8 is from my observation largely based in the L2 cache design with it's aggressive and speculative prefetch. AMD did not try to do this for K10. The L3 cache in K10 is complicated and it will be useful to improve finely threaded code running in all 4 cores of a L3 cache. But the K10 L3 cache does not include improvements that would be useful per-core like Core2 introduced.

[akaBruno]

Pi = 3.14 Dint take me long to figure that one out.

C2D will shine in some things. Phenom will in others. Which one will sell?

[cky2k6]

they both will... hopefully for amd, they can get more profit off these chips, as their volume is still decent, they just need to raise asp. as for me and probably the majority of xs that still has amd systems older than sam2, penryn ftw...

[JVguest]

Gary added some stuff:

There is no way that K10 with an IMC performs significantly differently with DDR2-1066 CAS4 (rare) vs DDR2-1066 CAS 5 (quite comon). There's barely any difference on Core 2 without an IMC and at 400 fsb!

[T2k]

Rubbish. There is no way the L2 cache was off in these tests. Hypertransport? OK, but it won't play a role for superpi and very little role for the per-core score in Cinebench.

Actually he could mean L3 and that indeed could be off. HT was planned to be off anyway IIRC.

[caligula]

Somebody please come up with some Core Celeron SuperPI scores@2GHz. Just to give AMD a hint what their work can be compared with
I am totally disappointed. Looks like AMD is willing to give the CPU market to Intel. Start stashing thousands of dollars for the next Intel CPU

[JVguest]

Rubbish. There is no way the L2 cache was off in these tests. Hypertransport? OK, but it won't play a role for superpi and very little role for the per-core score in Cinebench.

It's quite possible something was disabled. See early K7 benchmarks?
http://firingsquad.com/hardware/k7550preview/page9.asp

[Eldonko]

I hope performance improves a ton from this, because if not AMD will be gone.