Printable View
they used to have an 8 core on their roadmap I don't care if I have to wait till 2010. I will not be jumping on intel's first IMC chip.Quote:
Originally Posted by Blauhung
I didn't get 12.7 I got 12.2 with all my crap running it the background plus I was doing a large file copy and I'm using vista which is slower for superpi. if I wanted nice numbers I'd fire up xsos.Quote:
Originally Posted by Bellisimo
I think that this was a platform issue which has since been fixed.Quote:
Originally Posted by JumpingJack
3.2GHz = 12 secs = impressive indeed. My 3.6Ghz Q6600 gets around 13-14 secs.
Yeah --- my general opinion is that a) with early samples be it Deneb or Nehalem, while interesting, should be taken with some skepticism since pre-production engineering samples may or may not be actually representative of final performance, and b) in gaming Nehalem may prove to show substantial benefit in gaming code for some or even most games, but at conditions that most people play those benefits will not be noticeable.Quote:
Originally Posted by Shadowmage
The advantage to choosing nehalem would be more for future proofing, if anything, perhaps to pair with faster GPUs as they become available.
Thank you for reminding - I've read about this bug in a book, it's forgotten by most people, whereas AMD's blunder is quite recent. Intel's bug is probably of high(?) historical value but now the times, strategies, CEOs, product protfolios and architectures, i.e. all variables have dramatically changed.Quote:
Originally Posted by MrMojoZ
Are you sure? I assumed if the CPUs were so bugged that performance/IPC was dramatically decreased, the manufacturers would never allow leaks. Thus only give them to really tight-lipped partners who care about NDAs and do inhouse testing.Quote:
Originally Posted by JumpingJack
They're not stupid, they know how viral marketing works so why should they allow leaks of malfunctioning hardware showing them in a bad light?
All the publicly leaked conroe and penryn steppings were 99% representative of the end-product. Nehalem leaks seem also quite close to the final product (the numbers do not improve much if at all from stepping to stepping), albeit the memory controller was crippled for some time, but we can estimate the impact of improved bandwidth.
Do you think it is likely that AMD cannot produce bug-free steppings of a die-shrink?
If AMD is similar in that regard to Intel we know how K10.5 performs +- a couple percent due to an early leak. Surprisingly the preview shows ~8% better performance, the same IPC gains that behemoth competitor Intel achieved with their die-shrink.
I know this is not a proof, but if AMD cannot produce early bug-free steppings of a die-shrink, why should they be able to incorporate some magic that improves performance by 15% or more?
Hey,
I'm still packing a single core Opteron and an nForce4 board! I've been hesitating to pull the trigger on a new system for the last 2 years :p: so should I get a cheap E8400 and DDR2 now, or wait a little longer for i7 and DDR3 prices to stabilise, then go straight to quadcore?
Cheers,
Suman
Skulltrail + DDR3 = win :D
looking forward to see that plataform running
Go for the quad...specially with your upgrade cycle :p:Quote:
Originally Posted by Sumanji
lol, I'm not that bad... at least I have been keeping up to date with graphics (X800XL, 7900GT, 8800GTS, GTX260)!Quote:
Originally Posted by Shintai
Thing is C2D and DDR2 are so dirt cheap at the moment, and the motherboards/chipsets are nice and mature - like a fine wine! But with i7 I don't want to get screwed by crappy motherboards and scalped by pricing...
What to do? :shrug:
Q6600, 4GB G.Skill 6400 C4 PI, Asus p5q pro for roughly 400$ total, nehalem will be much more...Quote:
Originally Posted by Sumanji
P45+E8400 FTW!Quote:
Originally Posted by Sumanji
I've been itching for a reason to upgrade. Unfortunately, Nehalem doesn't provide me with one. Looks like the wait continues, while my saved money devalues.
I think the second generation of nehalems is the one to wait imo, tho i'm hoping this first generation are good enough to make me upgrade :D
You are gonna see both, if you care about money you wait till i7 ships then buy older tech at even better prices. Even if I was buying Intel parts I'd skip this launch completely. People with un-limited budgets of course have more options.Quote:
Originally Posted by Sumanji
Price depends where you buy. Motherboard...I dont think you get a crappy one either.Quote:
Originally Posted by Sumanji
I bought an E6600 on release day and a P965 Asus board. I payed about 330$ or so for the E6600. (It was 326$ in bulk from Intel) while bad suppliers took 400$+. And my board can still use 45nm yorkfields, OC like a charm and works flawlessly.
Shintai has shown actual experience here. Shintai, you should be ashamed for bringing facts into the discussion. hahahaQuote:
Originally Posted by Shintai
:eek:...but what about all those bugs some of these guys keep referring to? Guess that goes back to hope for the fanboi's, as first hand experience has dictated otherwise.
Money...moneymoneymoney. When you have the kind of money that Intel does with the kind of talent they do, coupled with the extreme amount of research and pre-launch tweaking they do you don't have those kinds of issues. This ain't 1985 anymore.
Can we have some of these to test if those claims are true?
Please. :D
or x48/ 790iQuote:
Originally Posted by Katanai
a yorkfield at 3,6ghz will do <13s. mine did 12,8.Quote:
Originally Posted by Loser777
the nehalem seems to have round about 600mhz advantage over the penryns. so a 3,2ghz nehalem should be as fast as a 3,8ghz penryn.
not that bad at all.
for benching it'll really be a killer, but i don't know if this will also be the case with gaming.
there was a CPU-Z SS on that post ??? :rofl::ROTF::rofl:Quote:
Originally Posted by Zytek_Fan
Well, you are assuming that either AMD or Intel purposefully allow leaks ... (the Hexus preview bench was pulled BTW) :) .... this may or may not be true... I can make an argument either way ....Quote:
Originally Posted by Jacky
Both AMD and Intel want to get functional Si (even if it is not completely performance ready) to system builds asap so they can begin their own BIOS, debug, design work. This means floating thousands of samples out and about, and even with carefully tracked samples ... some wind up in the hands of people who should not have them ... and then we see the leaked data. Intentional?? Perhaps, unintentional?? Perhaps ... I don't know but it would be conspiracy/paranoia of me either way to assume the worst.
Frankly, I think we start to see leaked data as soon as samples start floating out, and when there is no leaks ... means there are no samples. AMD hinted in their Barcelona followup that they thought they should do better getting samples to partners earlier for validation (I can dig a link if you need it).
On the other hand, it could actually be viral marketing ... who knows? It is reasonable to suspect, but unreasonable to conclude... this is what makes it viral I suppose :) :) :) you really don't know if it were an intentional plant or if it was an honest to God breach of NDA and some zealous whitebox builder with connections to Anandtech or Hexus.net floated them a sample and a board for a day.
In either case ... when Phenom came out in preproduction form, with fresh BIOSes and the debate/arguments from the pro-Phenom groups were "well, it is still engineering samples" or "BIOSes still need work, there may be more performance to extract before launch" ... I, personally, do not think this is a bad argument and it is not unreasonable to doubt the benchmark data until the actual end product arrives ... corollary, if a bunch of Pro-Core i7 people looked at the leaked, pre-production benches and want to argue that there may be more performance to eeek out in the final retail ... it is not an inappropriate opinion ....
Does this make sense where I am coming from?
Jack
It is the most famous bug of them all, it was also the most poorly handled from the PR perspective... and it was just as blown out of proportion as the TLB errata.Quote:
Originally Posted by MrMojoZ
jack
Well, I think AMD can produce a quality part that is more than fit for the market from a shrink or any other design rev for that matter. I have yet seen any CPU released that is bug-free, they all contain errata... typically, the errata are so minor they never even trigger, only occurring under highly specific circumstances. C2D has some > 150 errata listed last time I looked, some closed, some labeled no fix intended, etc. K10 has > 300 errata indexed, some (many) have been fixed are are not even documented in the last errata revision, others also have no fix intended, etc.Quote:
Originally Posted by Jacky
Both companies, Intel or AMD, when they make 'spokes person' statements related to 'expected performance increase' cloud the statement with ambiguity. For example, the 20% number AMD touts I absolutely believe in this case -- for many reasons, but my question is 20% in what? Sever based apps, perhaps just the SpecFP_Rate -- if this is what they are referring to then this 20% number will not transfer over to DT most likely because a good portion of that 20% comes from BW improvements moving to HT3 in the intersocket communications.Quote:
If AMD is similar in that regard to Intel we know how K10.5 performs +- a couple percent due to an early leak. Surprisingly the preview shows ~8% better performance, the same IPC gains that behemoth competitor Intel achieved with their die-shrink.
I know this is not a proof, but if AMD cannot produce early bug-free steppings of a die-shrink, why should they be able to incorporate some magic that improves performance by 15% or more?
However, if it s 20% on average for SpecFP, SpecINT, etc. (non BW apps), then it bodes well that 15-20% is expected for DT (as the non-rate Spec benches do not press BW to it's hilt).
Frankly, we just need to wait for benchs ... if you want my personal prediction -- Shanghai will see good improvements in Server, making it much more competitive to Harpertown over a wider range of benchmarks, and extending the SpecFP_rate lead. They will also get a good power improvement from the shrink. In desktop, ~ 8-10% clock for clock improvement over Agena on average with a select few apps getting as high as 15-18% -- gaming, around 5-8% if you just measure gaming code performance ... so very 'Penryn over Conroe' like relative increases but just a bit more in this case (Penryn was on average in the 3-7% range over Conroe IPC-wise).
Total guess on my part, so those numbers should not be construed as my trying to spout facts. ... they are totally my opinion.
Jack
nice now people actually rate there cpu on a base of pi calculations..... :ROTF:Quote:
Originally Posted by RaZz!
perhaps they should have supplied the score of cinebench multithreading and HT loving sw then it is 4+ york against 2.6 nehalem
HT3 is only implemented in new chipsets q1, so preliminary launch stays on current platform thx to large OEM :poke:Quote:
Originally Posted by JumpingJack
how you could read in my post i was talking about benchmarking. furthermore, superpi is a very good indicator on singlethreaded cpu performance. :rolleyes:Quote:
Originally Posted by duploxxx