They already mention Trinity in soon to be released drivers? man, I guess it is coming sooner than later :)
Printable View
They already mention Trinity in soon to be released drivers? man, I guess it is coming sooner than later :)
FX-4000 2module/4core is supposed to be in price range of the X6 (guess the performance) and way higher in price than any Llano. FX-8000 offer twice the raw power.
Could it be that SB-E and IB are chip close in performance and doesn't make any sense against just one chip (Bulldozer)?
HD7000 and Fusion Trinity revealed in the new catalyst 11.7pre
http://static.betazeta.com/www.chw.n...islands-nc.jpg
Bulldoze the SB, Devastate the IGP.Quote:
652,TRINITY MOBILE (9900),0×9900,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
653,TRINITY DESKTOP (9901),0×9901,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
654,TRINITY DESKTOP (9903),0×9903,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
655,TRINITY DESKTOP (9904),0×9904,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
656,TRINITY DEVASTATOR DUO (990F),0×990F,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
657,TRINITY MOBILE (9990),0×9990,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
658,TRINITY DESKTOP (9991),0×9991,0×1002,0×0,0×0,0×0,0×0,0×0,0×0,FUSIO N TRINITY,0×0,NONE,0×6,0×0,0×04,0×1
http://transformers2009leiva.hostei....evastator1.jpg
Forum new options put my crazy.
I thought FX-8000 is supposed to replace X6, with 8130P priced at $320 (vs $299 of 1090T).Quote:
Originally Posted by Nintendork
1100T is 189$ at newegg now...
chew, lol :D
re dealyed SB-E, I remember DR Who said that graphics on die will be starved by bandwith, but I guess he was surprised by Llano perf. I would love see his face when LLano was released :rofl:
Go back to when I joined the forum. :DQuote:
Originally Posted by XRL8
o_0 Why do those Trinity driver lines all make sense with their respective platform, except for one? lol
"656,TRINITY DEVASTATOR DUO"
MCM Trinity?Quote:
Originally Posted by Formula350
12 cores Interlagos Engineering Sample running @ 2.6GHz turbo core 3.2GHz
http://pastebin.com/VixAiT5u
BD turboing 4,2GHz
Quote:
Originally Posted by Mechanical Man
Thnx, finally something. Even if it is fake. :)
I see nothing there :rofl: apart from showing that turbo does work.
if multies are unlocked, its super easy to fake that and make it look like 4.2 is stock frequency
the only other thing we know is how long 2M superpi DOSNT take, lol (he let it run for 20 seconds before stopping)
Yeah, but the fact that it does 1M @ 4.2 Ghz / 1.4v is a good sign...(considering that was enough time for 1M to have passed)Quote:
Originally Posted by Manicdan
If we were to assume it's at all real, all we know is Zambezi at 4.2GHz takes >20 seconds for superpi 2M. While Thuban takes what, 45ish?
Correct me if I'm wrong but isn't this guy breaking the NDA just by showing this stuff? If he is, why the hell does he keep hiding everything and constantly running SuperPI like it's the mother of all the benchmarks.
Just run Cinebench and show the results, this is getting boring :rofl:
If he's only revealing it's existence, but not how it performs, is that breaking the NDA?
That's what I want to know as well.
And he did show how turbo is performing (not in numbers but...), maybe even showing the clocks/voltages might be an NDA break (just guessing here).
Dirty "24/7" gamer OS, no tweaks. Vista 32-bitQuote:
Originally Posted by Apokalipse
Excuse me for the dual desktop, I'm lazy today.
Click for larger
Thuban - ~39s
Attachment 117078
All NDAs aren't created equal. Plus, they might have lifted certain aspects at some point, perhaps to drum up some lost interest due to the delay.
Hard to figure out exactly why everyone has been so tight lipped. Not questioning why AMD wants it (all companies do), but why testers have been respectful enough not to, when it hasn't necessarily been the case in the past. Ya know? Is it perhaps that BD actually kicks enough ass that they want the initial "Wow!" to be that great, or is it that it sucks and nothing really worth leaking? :S
I think reviewers just dont have any bd's yet. And btw, if there is no review samples yet, then there is no NDA yet. So basically nothing to leak.Quote:
Originally Posted by Formula350
expect that he hit the stop button in superPI at 20 seconds.Quote:
Originally Posted by Apokalipse
For OBR NDA means "Need Demonstrate Asap!" :p:
There are NDAs for everything in life, so it's not just when something goes to a reviewer (in the computing world). There are NDAs between companies, and no doubt employees as well. I'm sure you know that there are NDAs for beta testing software, which is basically what is going on with hardware testers. I had to sign NDAs for the closed betas of Diablo II and then later on the Lord of Destruction expansion. If Chew is even just playing with one, I'd be very surprised if he hasn't had to sign one.Quote:
Originally Posted by Mechanical Man
Demonstrate the photoshop skills on how to hide superpi scores :ROTF:Quote:
Originally Posted by DarthShader
Actually all of our NDAs are created equal.Quote:
Originally Posted by Formula350
Why are we tight lipped? We always have been on CPUs. When you sell the majority of your product through large OEMs you don't want to disrupt their supply chains and stall demand.
From OBR blog:
Attachment 117116
The video is coming I guess.Quote:
Look, this is amazing! Fully stable in Cinebench R11 @ 4,85 GHz! SuperPi 1M - 5,1 GHz on AIR ... proof video:
I really hope this will be true! I want a 4.8GHz on Air :p:
AWESOME!Looks C5F+FX rocks!
http://1.bp.blogspot.com/-aJTldEb1Km...e11.png-oc.png
http://3.bp.blogspot.com/-roteoiOV-N...uperpi_oc4.png
http://youtu.be/haV93vh20Q0
I don't trust this guy one bit tho.
We know what was going on with his Llano overclocking, so maybe it's the same issue here.
5.1Ghz on air @ 1.608v! If this end being real, i can see BD retail hitting 4.6GHz on air @ 1.4v easily. But, what are the stock clocks?
I'm giving this RIG to my brother and I'm getting myself ready for BD very soon. I can't wait much longer...
I saw 1.58v on that Asus tweak utility.Quote:
Originally Posted by TheBreezyBB
EDIT: There's too much useful info be masked so I have no confidence on these, the guy always make a lot of things up, maybe even the frequency is wrong.
cmon AMD give us some info, when you were releasing Phenom II we had benches weeks before release, if you dont im going intel, and so are all my mates lol :ROTF:
hmmmm more interesting
5ghz stable sounds good
he is not the one man with BD....And what I know from second man ,-). Dont worry.
I know we have been there, but as you say again.So will I ;-)Quote:
Originally Posted by JF-AMD
That didnt stopped you from showcasing both zacate and llano performance earlier .In the before phenom era there was more info as well.Point one.
You also dont want to stop people from buying sandybridge now ,as its available now, and users dont have necessary info to hold off for BD (if theres a reason to). ? .Point two.
I would think intel has even more OEM supply and demand chains to broke, yet it doesnt stop them from releasing some information.Thats point three.
As for the obr scores.Theres no reason why BD shouldnt be able to clock to these frequencies.
What second man or you know? No teasing please:shakes:Quote:
Originally Posted by FlanK3r
I was speaking about reviewers. I think chew does lot more.Quote:
Originally Posted by Formula350
Such a Tease :(Quote:
Originally Posted by FlanK3r
Yes, hence it shows it takes more than 20 seconds. Like I said.Quote:
Originally Posted by demonkevy666
OBR's "benchmarks and OCing" are pure crap if you ask me. We know how true his Llano OC turned out to be, don't we?
And he said those frequencies were fully stable, well I don't see any benchmarking done on the video (which is the supposed proof).
And the "teasers" are pathetic, same old stuff every time, and we don't even know if the CPU is really running that frequency anyway.
I wasn't following him before so i don't really know. I'm taking all these with a grain of salt :P:
Cant make that assumption yet. If Phenom II did 5.1 @ 1.608v, you'd be looking at 4.6 at about 1.43v, so there is no "easily" there.Quote:
Originally Posted by TheBreezyBB
Well i'm talking from my own experience I can do 4.4GHz on Air @ 1.53 did not stress test and i won't be doing so. so, a 4.6GHz @ 1.4v ( +.02-.04 ) is feasible as far as i can tell :p:Quote:
Originally Posted by BeepBeep2
nothing is certain yet sadly :( I'm getting a CHV in a week or so + a 1100t to play with until BD arrives
Well i'm talking from my own experience I can do 4.4GHz on Air @ 1.53 did not stress test and i won't be doing so. so, a 4.6GHz @ 1.4v ( +.02-.04 ) is feasible as far as i can tell :p:Quote:
Originally Posted by BeepBeep2
nothing is certain yet sadly :( I'm getting a CHV in a week or so + a 1100t to play with until BD arrives
Phenom II "good" scaling = 100 Mhz per .05v +-.01vQuote:
Originally Posted by TheBreezyBB
...I'd trade you my 1100T for your 1090T, but I hope you know why ;)
happy B-day Beep:D
I would trade mine for A BD in blink ;) And happy BD to you :DQuote:
Originally Posted by BeepBeep2
I suggest a grain of the size of the meteor that wiped out the dinosaurs 65 million years ago. (estiamted at 10km diameter if you don't know :))Quote:
Originally Posted by TheBreezyBB
If he uses the word "amazing" in regards of an AMD product, then I am having an Admiral Ackbar moment.
Yep, happy Bulldozer to you Beep. :DQuote:
Originally Posted by TheBreezyBB
Haha, the goal four months ago was to OC an FX-8130P on this day...Quote:
Originally Posted by DarthShader
you can Thank AMD for ruining your Birthday now :p:
THANKS AMD, for RUINING MY SWEET 16!!!Quote:
Originally Posted by TheBreezyBB
Not really, I just hope its out by middle of August.
I'm looking at the same time frame. I'll probably buy an FX-8130p in early October since I'll be taking a 40 days vacation by the end of August :D I'll make sure I'm ready for it :DQuote:
Originally Posted by BeepBeep2
Happy birthday....but I don't think we'll see BD until September sometime.Quote:
Originally Posted by BeepBeep2
Certain rumors point to September 6th ([H]), but it could be sooner or later.
yey! more time to save up :D
while i wait
http://i.imgur.com/qveok.jpg
too bad i can't fit 3 4890's lol
http://i.imgur.com/QPPyc.jpg
I am quite sure you will not buy a 8130p in early octoberQuote:
Originally Posted by TheBreezyBB
Is it expected later than September? from what i know it's coming in Q3 this year and that means as late as 30th September.Quote:
Originally Posted by duploxxx
Happy Bulldozer to you BeepBeep2Quote:
Originally Posted by BeepBeep2
I find someone was an old friend of XS.
http://www.xtremesystems.org/forums/...=1#post2186052
No, what Duploxx says is there won't be part named 8130P in shops at all. Naming scheme for retail is different to what was leaked / made up previously.Quote:
Originally Posted by TheBreezyBB
Oh i see My bad :( . Thanks For the infoQuote:
Originally Posted by Lightman
If true, then Zambezi will be in my future. If not, then Zambezi will probably still be in my future because it's a lot cheaper than Intel offerings for the most part. Still, I hope AMD can return to competition in the high end market. The computer in my sig wasn't saved by my Cyberpower 1350AVR UPS. A lightning strike took out the PSU and motherboard (and possibly a lot more but I can't test them). I'm a hardcore PC gamer with 57 games just in my Steam account alone, 20 of which I play on a semi-regular basis. These gaming numbers are VERY important to me.Quote:
Originally Posted by Matas
Best wishes to AMD pulling a good gaming CPU out of its hat.
3 day old news...Quote:
Originally Posted by Matas
we take it easy on a sunday,:D over here
sorry. up here.
We cannot compare 2 different architectures clock for clock, but we should compare price, performance and power usage.Quote:
Originally Posted by Matas
If Zambezi is better performing at same power usage as i7 990x then Zambezi is better, its also by desing higer clocking at given TDP than i7 990x.
i am more interest on benchmark score of x8-FX
just wait...
Yes, indeed :D
@kam03(Sig):
Tease or real? Any indication, thumb up or down...?Quote:
AMD FX-8130P @ 4.80Ghz
Asus Crosshair V Formula AMD 990FX
Thx!
while we wait where can we get this!
this could be fun to play with
Attachment 117290
http://news.ati-forum.de/index.php/n...0t-aufgetaucht
Quote:
Originally Posted by Bolle
You will be glad you waited ;)
I thought there supposed to be no "8130P" model, right?
We should do the same with Olympians and sports teams too. Nah I'd say it's fair AMD used to brag clock for clock they could beat Intel. That was in the Barton days, 939 was great too. As consumers you should push for performance. A 6 core processor should not lose to the competitors budget dual.Quote:
Originally Posted by rintamarotta
How do u like this at FX?:)...:up: 2510 MHz at RAM
Attachment 117302
Sweet! Flank3r :eek: we want more :shocked:
That number doesn't say much without stating voltage. If it's within 1.4-1.5V then I like. :) Even without given voltage, I like the RAM clock. :)
So 3.7GHz base clock, or is that already with the +0.5Ghz?
Barton not only beat P4 willis clock for clock, it beat them alltogether, at lower price.So there was something to brag about wasnt it.For gods sake PIII`s often beat P4...Quote:
Originally Posted by Glow9
I know youre flamebainting Glow.But lets set something straight. X6 doesnt "lose" to budget dual intel.Yes ,it has higher IPC, so in some instances it will, but not overall, yorure just talking silly.
@Darthshader, its the same sample OBR has.Its OBR`s screenshot.B0 ES.Not any form of retail.So multi is irrelevant.
retail SKU... noQuote:
Originally Posted by DarthShader
that is interesting, i had the impression that they would never release the quad core thuban version..... seems like they now have enough broken cores??? could be a real high OC'er AMD wise.Quote:
Originally Posted by duron
It also stated 'overclock:34%' so it's 3.7Ghz at stock. Wonder if it is final retail or still immature ES.Quote:
Originally Posted by DarthShader
That should be easelly within 1.5volts at least in retail product.Quote:
Originally Posted by DarthShader
I'll leave the screens to Zambezi release but 4.8ghz 1.47volts, Noctua NH-D14 3-fan push-pull/push-pull config.
I doubt any ES's are yet 100% final retail products, there might be some differences but current samples ES's are way better than earlier ones.
Divide 269 by 200 and tell me what you get...Quote:
Originally Posted by undone
I can boot a 21x multi, and it will say 0% OC...
That sounds good. If final silicon can improve a bit on that, BD will be right where it needs, to be very competetive. :)Quote:
Originally Posted by rintamarotta
Maybe im off.But when i push my 1055T with 1.47v ,it starts to get HOT without of course extreme RPM`s on fans, and here we have 8 cores.I know that a lot depends on the design die area etc.Quote:
Originally Posted by rintamarotta
Could you tell us if this voltage is sustainable with not loud high end air cooling ? In some hard load of course like primex64.
rinta, can i move in with you for a few weeks? :D
We are talking 32nm vs 45nm here and a completely different core design...Quote:
Originally Posted by XRL8
For all we know, this CPU may tolerate up in the 75-80c range like i7 does.
Yeah it COULD.However when you look at past node shrinks, it has always come with lowering requiered voltage.32nm at 45nm voltage (at which it is very hard to keep stability due to heat) is more likely to be unstable.Quote:
Originally Posted by BeepBeep2
As for Intel comparison ,intel doesnt use SOI.And runs at lower voltage.I dont think correlation is proper.
When you take all things into consideration its less likely to be free of heat related problems than X6 at same voltage.Of course full load.Power gating should help a lot in more mundane tasks.
I personally deem "overclock" mhz wise ,when cpu is FULLY stable under prime for several hours.And heat is definietely an issue for most cpus.3x 3000Rpm Kazes arent real world "usable" too.Thats why im asking.
Nice, what about NB speeds? Will we easily 4000Mhz nb air cooled?Quote:
Originally Posted by FlanK3r
IMC setup is probably a lot different than what we see in Phenom II I would guess.Quote:
Originally Posted by Barr3l Rid3r
ok, but seem they still working w/ same clock base of 200Mhz... so NB should have compatile numbers. I saw it will have standard 3200Mhz NB clock. But what about headroom?
At this point in time BD NB clocks are only numbers for us. We don't know what was changed in MC in BD. Yes Thuban's like 2.6ghz NB speeds to use 1600mhz ram effectively. But BD MC can be more efficient and might not need those clocks. Or BD MC might be clocked higher than Thuban NB and thus there will be no need to clock it even higher. It might be the case that clocking NB will not give us any benefits, like todays HT link in thubans, which when you clock higher than stock even decreases performance sometimes.Quote:
Originally Posted by Barr3l Rid3r
So I suppose knowing how high can CPU clock go is important for us OCers, but we cannot say the same about NB clocks.
Its litlebit cooler than 1090T OC'd to 4.1Ghz with 1.47volts, will not say more.Quote:
Originally Posted by XRL8
Node shrink helps in that.
I can match that but with 8GB. :p:Quote:
Originally Posted by FlanK3r
hehe, nice bingo :)
Oh man flanker why didn't you tell us this was from OBR.:( Wasting my feelings at last. Perhaps ur still trust OBR at least sth, but I don't.Quote:
Originally Posted by FlanK3r
http://obrovsky.blogspot.com/2011/07...r3-on-amd.html
undone: in this case u can, believe me. If I had BD (or have :) ), until NDA could not show here :(...
I was trying to determine if thats a realistic 24/7 OC, if its the way you say, than its on the verge, but still doable without extreme measures.Thats good to hear.Quote:
Originally Posted by rintamarotta
Anybody know how its with sandys ? I know many people claim 5ghz and even up, but is that stable and is is sustainable is day to day usage ?
XLR:some SB chips are 5 GHz stable with good voltage (1.35V), of course, more of this can 5 GHz stable too, but with 1.45-1.55V. But dont forgot, SOI HK is diferent than Intels HK. I thinking, 1.45V at AMD will be still very safe for 24/7. More dangerous are temps than volts.