Deneb: sneak peek...
- Thread starter fellix
- Start date
That's too bad. I'm at about 1/2 that time @ 4.1GHz on an E8400. Lower voltage too. That voltage indicated is disturbing for a 45nm microprocessor, if true.
I so badly want AMD to be competitive on the CPU side again.
I so badly want AMD to be competitive on the CPU side again.
Meh. By the time it comes out, it will be competing with Nehalem (or worse / later)
I'd be surprised to see a 45nm chip from AMD in quantity before the end of this year.
Very much ditto. I'm sure we'll see samples, and we might even see a "release" towards the last few months -- but nothing in real quantities. Further, their 45nm technology is still pretty far behind Intel's 45nm process. Too many people look at Intels' 45nm processors and point to it being "45nm" without understanding that size isn't everything
Voltages are often a bit high for engineering samples, if that is what the readings are for. Stock volts could be lower.
Unless something significantly changed, it seems plausible that OCing will still require the heavy voltage bumps, though.
There seems to be some improvement per-clock when comparing the times to other SuperPi results for current Phenoms.
Unless something significantly changed, it seems plausible that OCing will still require the heavy voltage bumps, though.
There seems to be some improvement per-clock when comparing the times to other SuperPi results for current Phenoms.
Could also be caching hierarchy is better; the current line of chips has terrible latency issues on the L3. A 1M digits of Pi has a far better chance of being cache coherent on 6mb of shared cache, and if the timings are better, that could certainly explain the performance in that particular benchmark.
More benches are needed to determine how much better the computational efficiency is, not just for 19 successive calls of the same Pi calculation... Just IMO of course.
More benches are needed to determine how much better the computational efficiency is, not just for 19 successive calls of the same Pi calculation... Just IMO of course.
my understanding is that it's actually an iterative algorithm that takes 19 iterations to converge to the number of digits required. it's not just streaming out of cache.
http://en.wikipedia.org/wiki/Gauss–Legendre_algorithm
Ah, good to know. But having seen several benchmarks where SuperPI scores differently for no other observable difference except cache structure makes me still convinced that cache size matters.
From XtremeSystems forum member who did some testing:
Same architecture, same speeds, same caching methodology, doubled L2 cache size, different results. I think it's almost ironic that Intel picked up 12-13% performance increase by doubling cache, and here we are with the AMD chip netting almost the exact same thing from doing the exact same thing
I'm not saying that's the entire reason it's going faster, I'm not even going to say it's certain to be relevant, but I am saying that the SuperPi result is dubious at best in terms of what it can tell us about the performance of this new processor.
well sure, some stuff is going to be in the cache, but it's not just streaming out of cache over and over and over. it's not like you're running the same exact computation with the same exact inputs 20 times.
Thus far, I'll still say I'm not really impressed. 1.55v for 3.3Ghz? None of the benchmark scores look particularly impressive. And yeah, the thermals look OK for whatever cooling they're using, but I'm still pretty uninspired.
Let's see if they can get these shipping in anything resembling "quantity" before Nehalem comes along and squashes it flat.
Let's see if they can get these shipping in anything resembling "quantity" before Nehalem comes along and squashes it flat.
AMD 45nm Phenom first step review
Date: 2008/8/6
http://en.hardspell.com/doc/showcont.asp?news_id=3858
power usage looks really good, seems 5 to 17% faster than same clocked 65nm chip, depending on benchmark
Date: 2008/8/6
http://en.hardspell.com/doc/showcont.asp?news_id=3858
power usage looks really good, seems 5 to 17% faster than same clocked 65nm chip, depending on benchmark
Nice power numbers indeed.
But the performance still doesn't seem to be there; from what few tests I can compare from that site to other Q6600's, it still seems the Kentsfield is able to pass it clock-for-clock. That of course leaves an even bigger gap to the Yorkfield quads, and let's not even consider what Nehalem will do.
Still, for someone wanting Q6600-like performance at very low power, this could be your ticket. Nothing wrong with that stance at all; more and more people are looking for ways to save energy.
But the performance still doesn't seem to be there; from what few tests I can compare from that site to other Q6600's, it still seems the Kentsfield is able to pass it clock-for-clock. That of course leaves an even bigger gap to the Yorkfield quads, and let's not even consider what Nehalem will do.
Still, for someone wanting Q6600-like performance at very low power, this could be your ticket. Nothing wrong with that stance at all; more and more people are looking for ways to save energy.
Accord1999
Newcomer
Yorkfield is also very good in that department though:
http://www.behardware.com/articles/720-2/test-six-x3-x4-phenom-b3s-in-detail.html
Oh yeah, absolutely. Even moreso, the yorkfields are getting better performance with that decreased power draw. But they also come at a price premium versus the kentsfields...
So I should restate myself: for someone who wanted a cheaper, less power hungry version of the Q6600 (albeit slightly slower), the Phenom may be interesting.
(Well, we assume it will be cheaper, but if not... )Sorry but those scores are very dissapointing. My year and a half old E4300 @ 3,2GHz (400MHz FSB) takes 18 seconds for 1M test... though i indeed only have just 2 cores altogether But still...
I really wish AMD to make a huge comeback with something of Athlon4, AthlonXP or Athlon64 style...
Those were kickin' Intels butt badly in those days.
I really wish AMD to make a huge comeback with something of Athlon4, AthlonXP or Athlon64 style...
Those were kickin' Intels butt badly in those days.
AMD has always had relatively poor results in SuperPi, even comparing K8 to Netburst. Core2 just exacerbated the issue. There are much more important benchmarks to be concerned with. I'll take a Cinebench result over an SPi result any day.
