still waiting for retail
Printable View
still waiting for retail
I focus on the past because one data point doesn't make a trend. If this node and the next node both offer good gains in frequency when shrinking existing designs, I'd be inclined to say the trend has changed. Basing your prediction of a major improvement on a single data point when the last few before it didn't pan out the same way is being hopeful, not logical. Also keep in mind that I'm only talking about shrinks of an existing architecture and that I'm only commenting on node performance at introduction. AMD's processes always mature nicely. They just usually suffer a rocky start.Quote:
Originally Posted by BeepBeep2
I think it will not be based on core usage, but on power consumption.Quote:
Originally Posted by nex_73
Beep I know what you try to say, but AMD's 65nm process was bad, therefore it is bad to reference to it and extrapolate process improvements between nodes. On another hand 32nm process introduces Gate First and HKMG which somewhat balances things out. I doubt though it balances things enough for basing any clock improvement extrapolating AMD's 65nm->45nm transition. Besides we initially had 2.6GHz Agena to 3.0GHz Deneb! The difference you're referring to is for both mature processes. When BD launches 32nm SOI won't be mature enough to fully extract available performance. I expect close to 5GHz stable OC from BD style design on fully mature 32nm process after 2-3 design revisions. Initially I'm hoping to be able to clock 1 core towards 4.8-5GHz stable. All 8 cores @4.4GHz would be sweet for me as well. On air that is :)Quote:
Originally Posted by BeepBeep2
Bulldozer ES vs Core i7-990X gaming showdown
CPUs overclocked to 4 GHz, SLI Nvidia GTX 580
Have a look here, it is at the bottom of that page: http://pctuning.tyden.cz/hardware/za...-zari?start=11
Don't speak Czech or Google translate too gritty? Look here: http://www.maltrabob.com/BDvi7.html
Considering that we are still talking about engineering sample, the results are not bad at all in my opinion.
I wonder how many cores do those games use, if only use less than quad core then it's a good sign, because bulldozer module is not true dual core but somewhat weaker dual core, take the crappy ES bug especially frequency problem into account then Zambezi looks optimistic.
Ive got a feeling those games are gpu limited at these settings.Anyone has GTX 580 SLI to do a compare ?
Yes, but the ones above ~150 fps really start to show a difference.Quote:
Originally Posted by XRL8
AGAIN, OBR posted a Cinebench 11.5 stable screen at 4.67 Ghz. That thing has a GPU on the die...Quote:
Originally Posted by Lightman
If it were stars alone on AM3+ I'd expect up close to 4.8 Ghz with some chips. Add four more cores? Okay, lets go with 4.6.
...
Because they are clocked a lot lower :)Quote:
Originally Posted by Sn0wm@n
Wouldn't this be enough to indicate it's likely not a 'Dozer chip?Quote:
Originally Posted by XRL8
http://pctuning.tyden.cz/ilustrace3/...byte/bios5.jpg
1.4750V default, 2.0750 overvolted? :confused: :(
EDIT: Nevermind, I see what he did. Was showing max voltage for each. I'm going to assume that 1.4750 is also just the ES chip pre-set voltage to make sure it's stable. At least, I'm hoping that's the case.
I like how this came up after someone finally made a speculation about performance being 20% better then phenom II.Quote:
Originally Posted by chew*
when will allll this NDA be lifted!?
tomorrow :)Quote:
Originally Posted by duron
The NDA is under NDA. Good luck getting any information from AMD.Quote:
Originally Posted by duron
Where did you get this, today is Llano nda lift.As for BD, only krishna(or insert other deity here) knows.Quote:
Originally Posted by kam03
Ive been watching this thread and :rofl: at many of statements and have to agree with chew "I've got an idea. Stop speculating and just wait for the real deal. " I sure hope we wont need a BD to get us out of the hole you guys have dug with all the speculating and need bridge to make our way out.
If your enjoying it now, just wait till its released..lots of shouting for who was rightQuote:
Originally Posted by AMD_Freak
Not to mention the denial of fact from those who were wrong but refuse to acknowledge it.
Thats why i sigged two extremes ;-).Quote:
Originally Posted by Particle
I however think that the truth will lie exactly in the middle.
Erm... I pulled it out from my ass :DQuote:
Originally Posted by XRL8
Quote:
Originally Posted by duron
If you are referring to BDQuote:
Originally Posted by kam03
survey say's XXXXXXXXX wrong.
I give up.Quote:
Originally Posted by demonkevy666
If Stars is "greater than 6%" better than previous stars, I'd expect at least 15-20% from Bulldozer would I not? Otherwise, Trinity would be a total waste of time in using Bulldozer cores for a measly 4% increase (if it were 10% faster) to be made up by 100 Mhz in clock. Not to mention Llano is L3 free, who knows what performance benefits that core would have with L3? 10%? ...oh wait, I'm not supposed to be logical.
/end my posting in this thread
Tin hats.
Time to talk about some funnier stuff.
I took the time to modify some funny chops people did in another thread into a short narrative.
http://chew.ln2cooling.com/Simon%20s...ver%209000.jpg
I want not be bad, but...:-)
http://vr-zone.com/articles/sandy-br...012/12816.html
Yea, and Ivy Bridge surprisingly got pushed back to March 2012 too (read that a few days ago, before SBe delay was mentioned). Wonder if it has something to do with Intel using PowerVR graphics core, having problems getting it to play nice or as powerful. That's all I can figure given it's occured not long after Sabine platform launch, and then SBe happens damn near the same time as Lynx platform launched >_> Intel somehow misjudge what AMD's graphics were going to be capable of? :shrug:Quote:
Originally Posted by FlanK3r
Would it be over optimistic to think maybe Intels spys in AMD have bought back some info on BD performance so revisions are needed by intel so in Jan they can regain the crown that BD may take when it comes out? Any way, where are the lines drawn between hopeful > optimistic > over optimistic :rolleyes: