Link
If that is the case then INTEL will kill AMD by 2008. AMD sucks :mad:
Printable View
Link
If that is the case then INTEL will kill AMD by 2008. AMD sucks :mad:
saw that
*pukes
if that is the case it will be one of the biggest reason why i won't jump all over the CPUs like a rash
yack
no mention of memory config/timings used etc ???
AMD has never been a SUPERPI monster.
It wil be a 3d monster again as we were used to..
besides that it's 3secs faster at 200mhz less compared to K8
horseshyte man
K8 was absolutely demolishing Intels(netburst) until Ln2 was taken into consideration or until people starting seriously benching centrino (but not many guys did that really cept for a few loyal intel guys and a couple of hard core guys)
LOLQuote:
Originally Posted by dinos22
You seme to forget that amd was pushing to 21.00secs where intel was at already sub 18secs...:ROTF:
670 anyone?
I wouldn't rate how good a CPU is on how it does on superpi. Oooo goody, intel's CPUs can calculate pi faster, whoop-de-do, who gives a crap. I'm interested in the overall total performance. And I'm waiting until release to believe too much about it, a few benchmarks here and there with a lack of total system specs isn't very helpful to me.
Now I guess if calculating pi to the nth digit is your thing then whatever, do what you want.
B/c Netburst did well at SuperPI is the reason i discount SuperPI entirely.
It's nice as an initial check on system stability, but that's about it.
While I agree superpi doesn't mean jacksh*t, I must say it's not the only things apparently Barcelona suck in - look at Cinebench, look at almost any leaked numbers: they all worse clock-for-clock than current Intel numbers, let alone Penryn or after that.Quote:
Originally Posted by SparkyJJO
It sure looks dark for AMD. :(
what does a xeon@2Ghz with 1066/1333bus have for results?
If the 30K 06 is correct, then the future looks very very dark for Intel cause no penryn or nehalem will beat that LOLQuote:
Originally Posted by T2k
3dmark06 is anything but a CPU test.Quote:
Originally Posted by BeardyMan
EVEN the 30k '06 being true, Barcelona/Agena X4 @ 2GHz performing equal to Barton @ 2.5GHz on 1M PI just can't be good...
BUT we have to remeber that we can't confirm if that is true...
PS.: Is it just me, or the score really came from coolaler's site?
But that's why I'm waiting for release. All I'm seeing are some numbers here and there from different places, some of them aren't adding up, and we all know that some of the initial batches of K10 weren't good, so how many of these numbers are from the not-so-great chips? Makes me wonder. Once the thing is released then we'll know for certain, until then it is just a lot of speculation and I'm sick of it. Sept 10th get here already darnit!Quote:
Originally Posted by T2k
*EDIT* check this here: http://www.xtremesystems.org/forums/...&postcount=138
So it looks maybe these scores are done on the "broken" chips. There is hope yet. Sept 10th we'll know (or soon after).
If you get 11K in the cpu test, you have a monstrous cpu, period.Quote:
Originally Posted by T2k
GOD DAMMIT you ****:D :rofl:Quote:
Originally Posted by SparkyJJO
i was trying to slam my screen to get that bug :ROTF:
What part of the CPU does SuperPi use? It was my impression that Barcelona/Phenom will be have improved FPU capabilities compared to K8 along with a general improvement in per clock ability.
lol somebody fell for it already, I just stuck it there a couple minutes ago :p: :lol2:Quote:
Originally Posted by BeardyMan
I have to admit tho I almost smacked my screen not long after putting it there hehe
This SPi1m result is bogus.K10 @2Ghz performs better than that..EVT B0s or B1s that are bugged.
stupid bug... i am stealing it. i think the call should wait on wiehter or not amd will regain its crown
Somebody please post a 2.0 GHz K8 number, preferably AM2. It's so long ago I ran anything at 2.0 GHz I don't have anything anymore :)
BTW, it is likely/possible that this SuperPi score is screwed by the OS allocating memory in the NUMA memory bank in the other processor and/or switching processors during this run.
http://www.xtremesystems.org/forums/...&postcount=100Quote:
Originally Posted by uOpt
I get 45s on my 2ghz am2 cpu, but my timings are much more relaxed on the mem since I was having stability problems a little while agoQuote:
Originally Posted by uOpt
Or you just have a CPU which is well-tuned for those types of things - let's not forget 3dmark06's CPU tests are not a conclusive reality test, rather some hyped-up wannabe gaming tests built on Dlite and AGEIA's libraries, running in separate threads, that's all.Quote:
Originally Posted by BeardyMan
Wow, 39 vs 41 secs - that'd be REALLY-REALLY crap for a new architecture and for AMD stocks... :eek:Quote:
Originally Posted by Trunks
Relax,the results are not final:Quote:
Originally Posted by T2k
http://www.xtremesystems.org/forums/...&postcount=155
Thanks.
Now we only need a 64 bit Cinebench score for K8.
SuperPi really isn't too meaningful for CPU performance.
Well that changes the whole picture - thanks for the link; I'll put it into my sigline. :cool: :yepp:Quote:
Originally Posted by informal
good, if we're lucky we'll be seeing about twice the performance of that if amd's claim of 90% over k8 is correctQuote:
Originally Posted by informal
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.Quote:
Originally Posted by T2k
That's single-core. Of course a lot more happens for the across-core scores.
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 :DQuote:
Originally Posted by Jass
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 whileQuote:
Originally Posted by Jass
ēthat :yepp:Quote:
Originally Posted by BeardyMan
He says chipset, BIOS have changed plus HT activated as well as L2 cache is enabled now on the new chips.Quote:
Originally Posted by uOpt
All together it can easily redraw the lines - sound much closer to Intel, to tell you the truth.
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.Quote:
Originally Posted by T2k
I almost hit my screen also, thats a good one there...Quote:
Originally Posted by SparkyJJO
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
Gary added some stuff:Quote:
Originally Posted by [XC] flat-four
Quote:
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.
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...
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.Quote:
Originally Posted by T2k
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.Quote:
Originally Posted by [XC] flat-four
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.
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?
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...
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!Quote:
Originally Posted by T2k
Actually he could mean L3 and that indeed could be off. HT was planned to be off anyway IIRC.Quote:
Originally Posted by uOpt
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 :(
It's quite possible something was disabled. See early K7 benchmarks?Quote:
Originally Posted by uOpt
http://firingsquad.com/hardware/k7550preview/page9.asp
I hope performance improves a ton from this, because if not AMD will be gone. :(
:rolleyes: :rolleyes:Quote:
Originally Posted by caligula
that's what I'm worried about:(Quote:
Originally Posted by Eldonko
This is a waste of space!!
What? Netburst sucked royally at SuperPi. The only DECENT Netburst numbers were 4.5+ Ghz.Quote:
Originally Posted by Mav451
Quote:
Originally Posted by uOpt
I agree wholeheartedly. And maybe for the end consumer's sake, your predictions are slightly conservative.
oh yeah re the "L2 being disabled" business, Sorry but that is completely impossible. Super pi would be up in the 20min region for 1mb ( I have run it without L2 on a K8 before via a corrupted bios somehow managing to disable it)
Ya, agreed. Anyway, SuperPi is not a good program to measure multicore CPU. AFAIK, it's a single threaded program....Quote:
Originally Posted by SparkyJJO
..as well as the other 99% softwares you have installed on your everyday machine.Quote:
Originally Posted by davidletterboyz
To continue logically, we can deduce that AMD remains to be well suited only for multicore apps. In this case, who would buy it anyway? Enterprise users. And not even them, since in the enterprise world Intel has secured itself a huge share.
Think, people, think. Sometimes it's harder than everything.
i dont think its that bad, nowadays more and more software trys to utilize manycores. For example games.
Quote:
Originally Posted by dinos22
yes just as i insinuatedQuote:
Originally Posted by BeardyMan
what about equivalent cooling above ambient ;)
What I mean is that multicore CPU should be tested with multi-threaded program.... and superpi is not multi-threaded. :p:Quote:
Originally Posted by caligula
Quote:
Originally Posted by SparkyJJO
I just tried to but stopped before touching it since its a LCD. Canned air didnt get it either! :rofl:
sure a multi-core should be tested with multi-thread apps. But performance per core has to be good too.Quote:
Originally Posted by davidletterboyz
I posted on page 2 and repeat: This "K10" (wich we are not sure) @ 2GHz performs just about the same as a K7 @ 2.5GHz on PI.
Hope something is REALLY wrong...
I really don't care how this chip "SuperPIs". What I'm interested in is video encoding and rendering. I'll wait the ten days to see how this CPU performs.
Even SuperPi is a single core bechmark C2D still beats K10. What's the point single core or dual core, we can match core for core. If Intel's Quadcore can get under 10 sec superpi then AMD should also. :shrug:
When I use SuperPi on my quad , it doesn't even use any of my other cores. I don't really give a crap what superpi calculations crap is.
Here are my calculations in my quad core opteron system 2.6 though
4gigs of ddr2 667 ecc registered ram
http://i43.photobucket.com/albums/e3...titled-1-1.jpg
30% more clock than 2.0ghz K10, but only 5% faster. It seems K10 L3 has a nice effect in this crap test.
I really don't give a crap about this test either
yeah... the board looks impressive... lets hope K10 wont take emm down :up:
http://www.bit-tech.net/news_images/...0-733-news.jpg
I will believe any result when I see them posted on XS with an actual AMD K10 Cpu and with some screens with ram timings and all the stuff we usually use. Until than, I believe that that screen at the beginning of thje thread is just a hype or a teaser. as I am sure that the new K10 can't perform just like (or worst than) the old K8. Pretty soon will see what AMD's new arhitecture is all about....
2GHz quad K10 at 39 seconds? Hmm... look at the other K10 thread SPI 1M result, K10 quad core 1.8GHz, B0 step:
http://img128.imageshack.us/img128/3...220rs69uk5.jpg
Something is wrong with these tests - either one of them is not representative of the actual launch product by what I'm seeing. Why are they so unexplainable and contradictory?
I've already done something even worse, just to show how wrong this all looks. Heck a cheap old Celeron 256KB L2 which was murdered by Athlon 64 became close to the 39 second K10 time. :confused:Quote:
Originally Posted by caligula
P4 Celeron 3.5GHz 1x 512MB 2-2-2-5 SPI 1M: http://img250.imageshack.us/img250/2747/sp47dm3.jpg
I'm not even going to post the dual channel time it used to give. :ROTF:
The K10 32 seconds looks far more genuine than the 39 seconds result. This is too confusing and riddled. I'm leaving this labyrinth till September 11th. Adios. :)
Here is an old Opty 148 at 3100 for comparison.
The problem is SuperPi is really an irrelevant benchmark now. There's a wide range of CPUs that will score around the same.
uhuh. +1Quote:
Originally Posted by Monstru
when do HT 3.0 cpu's come out?...this is a future thing isnt it? or is ht 3.0 going to be utilised straight up my phenom/barcelona???:confused:
actually i dont know what ht is - it's a bus isnt it?
oh for cpu -> graphics...http://www.hardwaresecrets.com/article/19/3 - HT 3.0 - off topic:)
http://news.softpedia.com/images/new...the-Show-2.jpg
:spam:
says whoQuote:
Originally Posted by Zytek_Fan
it is one of the most if not the most popular benchmark at XS
what does superpi measure? the 'effect' of an overclock.
http://en.wikipedia.org/wiki/Super_PIQuote:
Super PI is single threaded, so its relevance as a measure of performance in the current era of multi-core processors is diminishing quickly. As the calculation times become faster and faster, Super PI is also becoming a rather poor stability test due to the need for a stability test to run long enough to stress the hardware.[citation needed] Many users are moving to Orthos or OCCT for their stability testing since those are multithreaded programs.
superpi is for single core cpu's
would not a multithreaded pi test be a step forward? for everyone?
the more programs that are altered to run in parallel the more benefits to the consumer (ie me)
i dont even know whether you can multithread a pi calculation..:lol:
you are mistaken if you think people are here to seek for 24/7 stability
XS in my mind is about pushing the limits in tech whether 3d or Pi
you guys keep confusing stability with benching. they do not go together. have you ever seen someone prime on subzero cooling....if you have they were just joking around put it that way.....it is just not what it's about
Pi can be run in multiple instances if you want to load all cores but who cares about loading all cores.............it's about Pi benchmark and you can even draw conclusions on rough 3D stability on basis of Pi clocks even........in AMD days dual 8m on dual core was good for any 3d bench, Intel i have not done a lot of 3d but seems about 100Mhz clock speed reduction on 32m stable
why is pi important? Pi is the best piece of software to determine how much RAM frequency and timings impact on performance of a system but it's true it does not necessarily translate across all benches and some 3d apps like tighter timings and other benches like higher frequency at looser timings ......
i start with every new CPU and RAM with Pi and quick 10m jerks in orthos to gain an idea very quickly where the system is at
24/7 stable systems people are soooo delusional about what are (for majority of people) nothing more than internet boxes and then the rest (bit of gaming, bit of video, folding or whatever you get up to) ..... you really don't have to go pushing the limits with 24/7 clocks as it just won't give you a massive advantage to squeeze out another say 200 MHz quad which is already stable at say 3.5GHz for daily use....all you will do is create problems in terms of heat, noise
:shrug: if you say so; you're the one with a closet full of hardware - not me :lol:
i am more interested in cpu design/architecture etc whether it is amd or intel;
i like seeing evolutions and changes.
bottlenecks and frequency limits :shrug: superpi is good for that?
superpi is one calculation; arent there many other types of calc/instruction?Quote:
why is pi important? Pi is the best piece of software to determine how much RAM frequency and timings impact on performance of a system but it's true it does not necessarily translate across all benches and some 3d apps like tighter timings and other benches like higher frequency at looser timings ......
if it helps you maximise optimum timings etc and frequencies yadada, well :shrug: if it helps it helps
you may as well use a sooped up single core imo. for superpi. - with your super cooling etc.
i'll ask again: what does superpi measure? raw compute power? no of transistors? effectiveness of cache? etc etc
does it measure overall cpu design or just one part of the cpu's ability?
i suppose you could design/taylor a cpu specifically to run superpi
geez you sig says otherwise :D:D:D:D:D :p:Quote:
Originally Posted by adamsleath
you got more than me :P :cussing: :lol:
some might say it's OCD (obssessive compulsive disorder) :doh:
yes i do suffer a bit from that :lol2:
All benchmarks which AMD losses become irrelevant...move along please.
The same could be said for Intel.
I've always said that, about all of us. :clap:Quote:
Originally Posted by adamsleath
Im waiting for some actual numbers myself then perhaps I will draw a conclusion.
Super Pi measures joined computational power of CPU / Ram / Nb.Quote:
what does superpi measure? raw compute power? no of transistors? effectiveness of cache? etc etc
does it measure overall cpu design or just one part of the cpu's ability?
It is highly dependent on CPU/Ram clocks, ram timings and northbridge arhitecture. It is a tool that measures the performance of CPU's clock per clock, using only one core. And if you want to se the overall calculus power, you can use wprime :)