Discussion in 'News' started by btarunr, Jan 13, 2011.
So 8 cores to compete with SBs 4?
As with anything else out there, doubt thats true. I'm not a processor expert. Most things I see, 2x the power, means a 50% or so increase. Performance gain usually isn't linear.
Needless to say, the CPU game has gotten real exciting!
Lol, I would laugh if AMD couldn't outperform the i7 980X with a 8-core bulldozer chip.
when it comes to things like encoding, its quite linear.
Actually the i7 950 came out in mid 2009. And while minor, their were several improvements to the design over that year. Doesn't make a big difference, but I take what I can get.
I hope you are implying 50% increase in performance > 50% faster. If you are not, then your statement makes no sense to me.
Sarcasm, it always falls on deaf ears in a serious discussion.
Server 16 core (8 modules) is slated to be the first Bulldozer on a market. For now it is expected to be accompanied by a 12 core (6 module) version on release day.
Desktop Bulldozers will be released soon after that and expected to be a 4 core, 6 core, and 8 core versions with a clock around 3.4 Ghz.
While I can vouch for the server chips coming out first and core counts, the clock speed and number of chips released at first are best guess.
I hope that is not true. As stated before, the i7 950 is faster than the 1100T in synthetic benchmarks. If I recall the 1100T sits a fraction below the i7 920.
So I am going with i7 950 8 cores (The system said it saw 8 cores so we go with that) vs. Bulldozer 8 cores = Bully wins by 50%. And knowing AMD that was a 3Ghz i7 vs. a 3.3 Ghz Bully so minus 10% due to clock speed and we get 40% better performance....synthetic.
Truth is, this means nothing. Without specs or details about the test setup and what the 50% relates to, this info. is pointless. I have to write this off as AMD's marketing keeping the buzz up about their upcoming chip since CES for them was all about the Bobcat APUs. Good job keeping us talking, but they better not disappoint us. There is a point that hype turns into a monster no amount of awesome can slay.
i would be more worried about what lga 2011 will have waiting. but i dont think that, if it beat bulldozer will be for much.
There are a few things that maybe linear, encoding is pretty straight forward. But multi tasking and gaming are far different.
I know some people out there are fine with spending $1000 on processors, I just can't, maybe someday that will be fine with me. But $300 and less is the sweet spot, and thats going to be the largest market too. Which seems to be where AMD aims, guess we'll see what both side release and at what prices though.
if AMD are quoting anything for this 50% claim, it would be exactly that. something that scales well in a linear fashion - a best case scenario.
you hope its not true? shit, a near 20% boost at the same clocks is exactly what you want from a replacement CPU.
If AMD can get 20% faster than current at the same clocks and 12 cores out... they may not win over intel fanboys and their single threaded games/benchmarks, but everyone else will be damned glad for the excessive multithreaded performance.
It have to be released in time, BS delays and crap will bring it down.
This Press note pretends to save their stocks from falling into an abyss
Salt mines stocks on a rise today .
Do you remember the Barcelona disappointment? Initial selective numbers showed a nice bump over Core2 for both int and fp, and we all know how it turned out to be in the end...
I smell more of the same. Hopefully to a lesser extent.
Well, looking at the number of initial responses, I foresee it succeeding.
this is great news
i knew that post on fudzilla couldnt be right, AMD wouldnt bring back FX for same as i7
I want the fact it was a best case scenario to not be true. I want a 50% boost when compared to Intel when AMD is playing in their ballpark by their rules. I hope 50% turns out to be a single threaded test to prove AMD improved the clock for clock fight they have been losing for so long.
Performance improvement always happens. I want a giant leap of improvement to put us back in a head to head fight. Then maybe AMD's new CEO will buy a marketing department and fight for real.
There are no more delays. The chip is done and being manufactured. It is all waiting for real numbers, real benchmarks, real samples, and then release. I can speak for everyone when I say to AMD, "We are frustrated and need a release ASAP. Move the release date up if anything."
FX is coming back. AMD was serious about that announcement.
Barcelona is a long time ago. Lately I would say AMD has been delivering on their promises. I have been extremely happy with all my AII and PII's so far.
I'm very interested to see how it overclocks. Given Intel's choice to block overclocking w/o K-series chips I suspect AMD might have an exciting value lineup.
me too, i was really considering upgrading to Sandy Bridge, but wanna hold out to see if AMD's claims for Bulldozer are true. If it is, then prices will drop even more i hope.
Whatever apps AMD using for the presentation, it wouldn't amaze me.
What amaze me is that if AMD's bulldozer run @1.8ghz on that presentation.
If that is true, that is some serious shit coming from AMD.
And since AMD plans to revive the FX brand again, it's coming back to the glorious days of Athlon64. 1.8ghz A64 3000+ OCed to 3ghz.
I swear that I saw somewhere showing an Integrated Electronics Corporation Core i5 750 Central Processing Unit only *just* getting beaten by AMD's 1100T. I would seriously doubt these new CPU's would perform even near as good as the 980X, let alone mid-range Sandy Bridge CPU's, let alone the high-end socket 2011 Sandy Bridge CPU's. A fair comparison, should be comparing two different 8-core CPU's in my books.
Hype from AMD has always tended to lead to disappointment. As Becker would say, no expectations, no disappointment. +1 to the skeptical kitty.
"AMD sure drag their hole, 'The Future is fusion' motto they have been using in their logo for ages now has still bared no fruit to my knowledge. Intel beat them at that too with the release of their Sandy Bridge processors, before AMD could even release their Bulldozer CPU's. Here's to hoping though, I want cheaper CPU's."
Skeptically 50% faster, at a 50% slower release rate.
"Everything beat the Intel CPU's (Pentium 4's / Pentium D's which were using the horrible Netburst architecture at the time) when AMD's FX series were out. Then if you rightly remember Intel's Core 2 Duo's came out mopping the floor with them and have been miles ahead since.
The FX's I'm betting will struggle to keep up with the current 'high-end' Intel Socket 1366 i7's, since the six-core Phenom's (proof as per mentioned on here elsewhere) are only about as powerful as the top 'mid-range' Intel Socket 1156 i5's. So I doubt they will ever reach anywhere near being able to compete against the new Intel Sandybridge architecture Socket's 1155 (the new i5 and mid-range i7 socket) and 1356/2011 (the new high-end i7 socket TBA). For that to happen they would have to have one hell of a trick up their sleeve so to speak, cos they would have to jump above like 3-4 series of Intel CPU's in performance to be able to claim top dog. Also doesn't make too much sense, having their logo being 'The Future is fusion' for quite sometime now, and Intel release their integrated graphics CPU's before them."
Hmmm, perhaps a ploy by AMD to hold back technology from the public so they can seem as though they are still in the game at later dates as they see fit in accordance to the competitions various releases. I sure hope not. I don't want to start a conspiracy, but if these suggested improvements are indeed true, they sure randomly made one hell of a big jump out of nowhere.
you cant compared bulldozer with phenom II it is a completely new architecture. the phenom II is K10.5 pretty much a 10 year old architecture so it held up pretty well.
Here's to seeing what it will be like on release then.
Separate names with a comma.