hope its true, with the supposed new stepping in q1 2012 giving better ipc along with AMD's recent history of process improvements (6ghz -> 7ghz on 45nm) this thing might not turn out to be worse than amd athlon x2 after all ;) 10ghz by 2013? :D
Printable View
hope its true, with the supposed new stepping in q1 2012 giving better ipc along with AMD's recent history of process improvements (6ghz -> 7ghz on 45nm) this thing might not turn out to be worse than amd athlon x2 after all ;) 10ghz by 2013? :D
AMD on Twitter has announced that it will reveal a secret tomorrow at IDF
"Our adrenaline is pumping from holding in our BIG secret! We'll be spilling the beans tmrw at #IDF2011. Stay tuned!"
http://twitter.com/#!/AMD_Unprocesse...80895834128385
The preliminary Bulldozer benchmarks are making me nervous. I can't imagine AMD would design a chip that loses to it's previous lineup. An 1100T crushes BD if those numbers are true LOL. Why would a site risk their credibility by releasing such dismal numbers.
How is this netburst? Just because it is designed for high clock speeds doesnt mean it will be a "netburst". Much of the shortcomings of the Netburst architecture have been addressed in the BD architecture so claiming BD to be a failed product before we even know real numbers is just silly. All the P4 was was a high clock architecture with no IPC, Bulldozer is a high clock architecture with a decent amount of IPC to go with it and lots of cores on top. Whether it will truly be good or not we will have to wait and see, but just because it is designed for high clock speed doesnt mean it is going to fail hard
Where are you getting the low IPC part from?
i never said failed, i was just pointing out the really high overclock they supposedly got. and if somehow BD can get like 6ghz on air, then no one will really care about the lacking IPC if they can just OC it past SB. (i doubt 6ghz on air will happen, but we dont know)
Attachment 120017
Tomorrow is INTEL DEVELOPER FORUM and AMD is going to reveal something there?
Hahhaa, AMD must have something HUGE planned.
I hope it's not just an 8GHz BD on LN2. I mean that's great for enthusiasts, but it's not like everyone here is cooling their rig with a LN2 pot.
They must have just ONE good chip to play with. Cause we sure as hell ain't seeing anything!
Kindly restrict your trolling to Intel section... We're here in "AMD" section to read some bits and bytes about the new processors which are to come out sometime soon... If you've nothing worthwhile to add, please do not be a nuisance.
To admin and other staff: Sincerest apologies... but i guess someone had to herd the trolls to the barn...
On topic. Bring out the news already... why would you deprive us of our sleep... 8+ ghz is good, and i'll conclude on LN2. However, if it manages even close to 6 on air/ water, it would be quite something alright. Personally i don't overclock much. I do love how AMD platforms are better specced for the same money, and by a mile, and if you're on a budget... unless you have some specific use in mind, AMD is the obvious choice, which i recommend to people who ask advice on such matters. However, if anyone from AMD is paying attention, in India, your dealers resort to price gouging (on purpose... hint, hint!), thereby positioning Intel's processors are better value. Sometimes, AMD is plain screwing up... which can't help you gaining market-share in any manner. When i bought my old system, C2D e6600 & e6750 were priced cheaper/ same than a X2 6000. Just saying...
On the architecture... there seemingly are bottlenecks that people are pointing to, but i guess, programming has not even started to use all the cores on a processor much, leave alone hitting a bottleneck (excluding specific server usage). All in all, i expect nothing less than 10-20% due to improvements in the prefetch, and some other improvements, at the same clock. If there's anything less... i'll guess that will have to be explained as some snag (aka TLB of Phenom), which will explain Charlie's statement WRT improvements in Q1 2012, with a new revision. I also think that it may have something to do with specific patches to be written for the OS to enable usage of all the improvements brought in by the new instruction set(s). AMD did write one such for Linux in the July this year. Next year, Win 8 is to be hitting shelves, which may bring a new SP for Win 7 & Vista (may be). Too many questions...
Sorry mate... it is getting a handful to distinguish between trolls and harmless banter :confused:
Once again, apologies for ruffling feathers :(
I did not know about a section called "for sale." :eek: my activities here are limited to news/ cpu and gpu sections :P
It is obvious unless AMD screws up big time like TLB on Phenom, BD will see some gains (at the same clock) in the range i mentioned. Intel also gained a whole lot in Nehalem and now SB, so why shouldn't AMD with BD? Now, precisely how much, it is anybody's guess. I merely posted my views on all that i've read so far.