Printable View
The older one is 1.4 GHz, not 2.1. Also, it only has 16 GB RAM, however I have no idea how much that affects those benchmarks.Quote:
Originally Posted by informal
Edit: Analyze this - http://browse.geekbench.ca/geekbench.../484217/317797
It's 2.1 Ghz.
3000 posts, quite a milestone
OK, we had some pictures of early AM3+ socket, and there was no way you could fit BD into Am3 socket:
http://www.freeimagehosting.net/uploads/32f02b7d8c.jpg
you can see there is an extra hole in AM3+ socket, left hand side, left lower corner.
Of course maybe BD chips dont need that extra hole and AMD just cut it just for kicks. But where is the fun in that? :D
1.4Ghz is the idle state of the cores... Read the OPN of the early ES,the base clock and Turbo are in the OPN : ZS21245TGG44_31/21/2_2/16. Base is 2.1Ghz and Turbo is 3.1Ghz,part is 16C.Quote:
Originally Posted by Mats
As for your second link,you are comparing an ES Opteron platform with a retail one. I used ES vs ES comparison.
That's the point, and that's exactly the story with AM2+ and AM3. Those pics aren't worth anything without knowing what the pins looks like on Zambezi.Quote:
Originally Posted by muziqaz
By your reasoning, the same pic tell us that AM3 CPU's wont fit AM2+ because of the extra hole in the lower left corner.... ;)
Yes, and my point was that the "huge" difference between the two ES isn't worth anything as long as it's slower than the last gen.Quote:
Originally Posted by informal
not sure i like seeing the ES lose to a retail 24core
Exactly, that's why I can't see why it's even worth posting the ES - ES comparison.Quote:
Originally Posted by Manicdan
It's worth because what we have seen so far are only ES results of both Opteron and Zambezi. And the difference between different samples running on different motherboards is massive. So ES performance varies a lot and is not final. That was my point.Quote:
Originally Posted by Mats
Also,do you believe 16C Interlagos will loose to 12C Opterons in any of the tests ,especially in single thread test ?
No way. I just think it was worth pointing out to those who aren't used to geekbench, me included, that these are not results that are close to the final product.Quote:
Originally Posted by informal
You didn't even call the newer one an "ES", making it even more confusing.
Well I assumed people are able to read. This is listed when you open the first link in my post : AMD Eng Sample, ZS232545TGG45_32/23 .Quote:
Originally Posted by Mats
I was talking about what you wrote:
Quote:
Originally Posted by informal
So? Click on the links and you will see ES markings on both of the links. Why are you beating the dead horse? What I wrote in my short post is not important since I listed no scores. Click on the links and you will see the system configuration and exact numbers.
yeah i thought at first you had listed a retail chip, which is why i clicked, if i knew it was just another ES i honestly wouldnt have cared.
You expect to see numbers for retail chip before it is actually launched?Quote:
Originally Posted by Manicdan
It's dead?Quote:
Originally Posted by informal
Edit: I'm just kidding. But It seems I'm not the only one who misinterpreted at first.
That's easy to believe when Cray already got their first box of chips. Not that I think it will be available that same day, but the question is when.Quote:
Originally Posted by informal
i only expected there was a reason you didnt call it an ES, and since we all saw a box of chips go out, its not unbelievable to see retail samples out there. these things are still WAY too slow, and it would be nice for some good news for once.Quote:
Originally Posted by informal
well i thought too that maybe it was a retail since they started shipping earlier this month(7th if i remember correctly)Quote:
Originally Posted by informal
i wasn't lookin @L2 cache ever lol the nb can also effect core to coreQuote:
Originally Posted by TESKATLIPOKA
bulldozer has a lot of power saving features compared to phenom IIQuote:
Originally Posted by TESKATLIPOKA
for what Im curious now, how good will be OC at uncore :).
I'm just curious, i knew that 4 models where suposed to be launch not three.
The missing one is FX 4xxx quad core. May that his performance not worths to launch comparing to X4 955 which is at 112$?
Or may be AMD didn't even make a derivate design - FX 4xxx quad should be more a BD 8 core split in half.
Because the 6core FX 6xxx it's clearly just an 8core with 1 module broken.
the missing 4100 sounds like they are not making enough chips to need a new bin, if they are getting chips that have 2 broken modules that can only be sold as 4100s, then they should be around, but if there are just not enough of them it could be too soon to include them. or they could be worried about people unlocking them and not buying the 8100s initially. or yields are great and there is no market for half broken chips cause all the bad ones are still good enough for 6100s.
Don't compare Solaris Geekbench results (the Opteron 6176SE system) to Linux (Interlagos system)
The variation in scores between Windows/OSX/Linux/Solaris can be huge with Geekbench. The scores shift by a not insignificant amount with each minor update to Geekbench too. I've seen a massive increase in the Blur Image/Sharpen Image tests on K10 going from Geekbench Win 64 2.1.13 to Geekbench Win 54 2.2 (thanks to the new compiler used).
Geekbench scores across different operating systems and different Geekbench releases should not be compared.
demonkevy666
Thats true but what are you trying to say with this? That they have some power saving feature for cache enabled non stop and thats crippling the performance or something like that?Quote:
bulldozer has a lot of power saving features compared to phenom II
Yes the 4core would be an Orochi die with 2 deactivated modules, comparable to the the Phenom2 5xx line.Quote:
Originally Posted by xdan
There are several reasons for holding it back:
a) Sales: Too much Phenom2 X4 & X6 inventory, AMD wants to sell it off first.
b) Marketing: Emphasize on the "high-end" 6core and 8core models, the FX4 might influence the image negatively, the prices are already quite "cheap".
c) Optimize ASPs: People who are interested in BD and waiting for it, should buy at least a more expensive FX6 and not a cheap FX4.
I for my part will wait for FX4 ^^ Cheap and I assume easily OCable@5GHz, because there are only 2 modules ;-)
That's enough for me.
I don't know. I'm also confused by there being two 8-core. if this chip has a short life on the market, and they're niche chips already, and both are unlocked, so both overclock the same, and... I just don't get it. MAYBE after respinning 10 times they've found yields are so good to not warrant selling many broken chips.Quote:
Originally Posted by xdan
Would be interesting if they allow you to run one core per module, I imagine that would mean the 8 core could become a very nice quad.
Well that's for sure one of the reasons they're holding back!!Quote:
Originally Posted by Opteron146
A 1055T for U$S 195 and a 955 for U$S 146, those CPUs have never been as cheap as they are now here in Argentina, every AMD cpu had their prices cut... it seems like they're almost giving them away! I'm sure it's the same all over the world and especially there in the USA. I'm looking forward to october 13th! (or mid october, or whenever they decide to launch these new CPUs). My E7300 + P35-DS3P are quite overdue for a complete upgrade... that mobo is at least 5 years old, it got enough parts upgrades already...
Just because DonanimHaber doesn't tell us the price of the FX4 it doesn't mean it's delayed.. it isn't the most reliable source in the world.
AMD sets prices after how their chips perform compared to the competition. Not after die size. Production cost isn't part of the pricing. If their chips cost 50$ or 150$ to produce won't change that. Even if they cost 150$ to make they can't put it at 400$ if it performs like a 200$ chip. It wouldn't sell at all. If a chip is to expensive to produce to fit it's price range it will simply not be produced.Quote:
Originally Posted by muziqaz
Compare with Radeon vs. Geforce. Their pricing reflect performance and functions, not die size. nVidia had lower margins per graphics card as a result. The GTX 280 was extremely expensive to make since it was a large chip, but mainly because of the yields being so bad. But they sold it at the same price level as they sold 8800 GTX at it's time, and cheaper than 8800 Ultra.
Hmm already forgot the now offline listings last week in the online shops? There were only the FX6 and two FX8, not a single FX4. Exactly as DonanimHaber tells now. Therefore, the whole story seems to be pretty accurate to me.Quote:
Originally Posted by Mats
Here ya go....we can vote our speculation now....LOL
Poll: Gauging Bulldozer's performance prospects
It's hard to forget something I've never noticed. Anyway, as you say that list is gone now so I have no idea how accurate it was.Quote:
Originally Posted by Opteron146
Edit: A bit like saying that there will be only one Interlagos model at launch, because the same site lists only one model:
http://www.shopblt.com/cgi-bin/shop/...er_id=!ORDERID!
Dunno if it has been seen but read this a few minutes ago:
AMD Bulldozer CPUs Launching October 12th w/ Prices Starting at $175
For that low of a price i see a 8150 in my future instead of a 4100
Similar minds think alike :). I'm waiting for the cheap FX4xxx to put it in a cheap AM3+ board and drive it to ~5Ghz on good air cooler. I doubt I will need anything faster than this and with help of new power management and updated K10stat I expect to run it at super low clock and low voltage for the most time (when I don't need that much of compute power). It should be massive step up from my trusty K10 @ 3Ghz which BTW is doing great ,knock on wood.Quote:
Originally Posted by Opteron146
Not sure it's the best choice for you if 5GHz is your goal. It's most probably the same Orochi die with 2 modules disabled. Usually not because those modules are broken, but that the die has subpar electrical properties, so that with all the modules enabled it wouldn't fit in the 125W TDP class. Hence the realtively high TDP.Quote:
Originally Posted by Opteron146
EDIT: But, who knows... Perhaps it's still capable of 5GHz, but at a higher voltage, which results in a considerably higher power draw (compared to that if you get an FX 8150 and disable two modules manually or set only the 2-module turbo clock high).
BTW, just heard GloFo is turning most if not all of their 45nm lines to 32nm. Wouldn't think they will do that if BD was such a disappointment...
$146? Thats expensive! I just bought an 955 for $105 2 days ago. New from newegg.Quote:
Originally Posted by Dr. Vodka
So last time AMD spoke about Zambezi's availability 3 months in anticipation (with the result we all know) but now we would be 3 weeks away from launch without any official word from them.
Doesn't make any sense, at least to me.
That doesnt mean Zambezi is doing great though, thats just progress. No future things from AMD will be made on 45nm, its all moving to 32 and then on lower. They really only need a couple places to stay on 45nm for other customers who still want that, but the majority of their business is from AMD and AMD has moved on to smaller designs all across the board.Quote:
Originally Posted by dess
Yeah well he is in the Land Of Silver, not USA..Quote:
Originally Posted by EniGmA1987
Right, but if Zambezi is such a disappointment as some may think it is, then there would still be a higher demand for Thubans, even Denebs.Quote:
Originally Posted by EniGmA1987
Otherways, there were news on Phenom II's going EOL shortly. Only Athlon II's remain a little longer, they say. (I mean AM3's, not those on FM1, made out of Llanos with a disabled IGP.)
http://www.criminalcafe.com/showthre...ll=1#post64953
Don't know if you guys know him. But he's a very respectable brazillian enthusiast.Quote:
Originally Posted by PcCI2iminal
He says "FX8150 is faster than any variation of the preview uarch (faster than any K10) and in some few cases its equal or faster than the i7 2600K. But on avarage it's slower than the 2600K, that's why the price droped to U$244".
he's from ArgentinaQuote:
Originally Posted by EniGmA1987
In Brazil a X4 955 costs you R$280, wich is about ~U$155
very expensive, isn't? :D
I wish we had your prices here! U$S 350 for a 6950 when you guys pay U$S 100 less. That's the way it is down here.. Anyway, there have been noticeable price cuts on AMD's entire CPU line, as someone said a few posts above me, they're probably trying to sell as many PII cpus as they can.. while they iron out any last minute bugs, etc.Quote:
Originally Posted by EniGmA1987
------------------------------------------
So, in some cases it's equal or faster than 2600k but on average slower... that sounds like it's quite close! Nice! Just to think that AMD went from not being competitive at all to being up there with intel's best is nice to hear. Oh well, time'll tell the truth... It's good to hear this, plus movieman's words and the like... And most importantly, what seems to be competitive performance at an attractive price (if what's on the rumor mill turns out to be true in the end). I'll gladly wait a few more weeks...
Yeah he's a member here too: http://www.xtremesystems.org/forums/...98-PcCI2iminalQuote:
Originally Posted by Lokinhow
i sure freakin hope so i7 2600k has been out for almost a year! no wonder there pushing piledriver out soon after... there gonna need that arch updateQuote:
Originally Posted by Lokinhow
after some bench leak
1055T in my country price increase
seems like your Vendor knows something....Quote:
Originally Posted by PatRaceTin
Yes.. he knows how to make a quick buck off of those who don't know better.Quote:
Originally Posted by duron
At the same time, EUR/USD went down:Quote:
Originally Posted by PatRaceTin
http://finance.yahoo.com/q/bc?s=EURU...=on&z=l&q=l&c=
;)
In Italy 95 eur, ~U$130.Quote:
Originally Posted by Lokinhow
However, is it a "Sempron BD" a dream? 2 cores BD for low budget config ...
Fottemberg: There is no 2-module BD die, so it would need to disable 3 modules on an Orochi die to have a one module active only part. I don't think it's going to happen.
There are Phenom II's and Athlon II's for that purpose, until EOL'ed.
Otherways they cover mainstream and budget with Llano now and later with Trinity.
Do peeps not like Legit Reviews or something? I posted this yesterday and just about everyone ignored it! :shrug:
AMD Bulldozer CPUs Launching October 12th w/ Prices Starting at $175
It's a repost from Donanimhaber news which we knew already. But it won't hurt I guess.
Damn, oh well! I didn't see it :D , sorry for re-post; again!Quote:
Originally Posted by informal
There are so many friggin pages now that it's hard to read it all to see whats been posted.
yeah i love when someone says, "we mentioned that last week, keep up!" only to realize that means 20 pages ago
anyone else hate how good AMD is at keeping secrets.
True. :up:Quote:
Originally Posted by Manicdan
... AMD Employee loses a Bulldozer prototype in a bar ... :ROTF:
.....now we're talkin!! Name the bar and I'll be there! LOL
To leave a CPU, or to pick one up?:eek: You have one? :ROTF:Quote:
Originally Posted by Canis-X
Well I hope the price for my hometown won't increase so much ......:shakes:Quote:
Originally Posted by PatRaceTin
LOL.....If I actually had one, I wouldn't be sitting in on a forum killing time. I would be benching/gaming and doing my best not to tell anyone anything that I wasn't supposed to.......twould be extremely hard too! :rolleyes:Quote:
Originally Posted by Mats
I respect that, what i hate tho' is that they remain silent after missing their deadline so badly.Quote:
Originally Posted by Manicdan
Uh-Oh....
http://semiaccurate.com/2011/09/22/a...-bergman-gone/
http://techreport.com/discussions.x/21697
http://hardocp.com/news/2011/09/22/a..._cloud_mystery
http://www.amd.com/us/aboutamd/corpo...k-bergman.aspxQuote:
AMD Fires Rick Bergman Under Cloud of Mystery
Rick Bergman, Senior Vice President and General Manager at AMD has been fired according to HardOCP sources. AMD has given no reason as to why the well seasoned Bergman was leaving his position. It was told to us that Bergman had interviewed for the just-filled CEO position months back at AMD and had been turned down. Since then we have been lead to understand that AMD has lowered the boom on Bergman for not managing the GlobalFoundries relationship properly. Sources tell us that GlobalFoundries is simply not up to the task of supplying AMD its needed parts and Bergman is first in line when it comes to the responsibility of making sure AMD is sourced properly. We understand that Bergman will be the first of the dominoes to fall.
Rick Bergman is senior vice president and general manager of AMD’s products group, with responsibility for delivering AMD’s computing platforms and managing the graphics and microprocessor product development teams.
I first met Rick a bit over 10 years ago. Rick was a hell of a nice guy and we wish him the best.
Quote:
Executive Biography
Rick Bergman
Senior Vice President and General Manager, AMD
Rick Bergman is senior vice president and general manager of AMD’s products group, with responsibility for delivering AMD’s computing platforms and managing the graphics and microprocessor product development teams.
Bergman’s previous role at AMD was senior vice president and general manager, Graphics Product Group (GPG), which he held from October 2006 to May 2009. In that role Bergman was responsible for the worldwide management of AMD’s discrete graphics products, including the company’s industry-leading ATI Radeon™ family of graphics processors for the desktop, workstation, notebook, multimedia and game console markets. In that capacity, Bergman oversaw business operations, engineering and marketing, as well as iconic achievements such as the first DirectX® 10.1 GPU, the first TeraFLOPS-class GPU and the first GPUs on 65nm, 55nm, and 40nm process nodes.
Bergman came to AMD via the acquisition of ATI in October 2006. He joined ATI in January 2001 from S3 Graphics, a division of SonicBlue Inc., where he served as chief operating officer.
Over the past 15 years, Bergman has served in a number of marketing and management roles within the North American technology industry, including vice president of marketing at Exponential Inc. and marketing manager at Texas Instruments, Inc. He began his engineering career at IBM Corp.
Bergman holds a Bachelor of Science degree in electrical engineering from the University of Michigan and a master’s degree in business administration from the University of Colorado.
<duplicate post.....sorry>
Well, someone was eventually going to have to take the blame for the incredibly long delays.
So it seems. Perhaps the problem all along with BD was not performance issues, but supply issues. Global Foundries can't seem to do anything right lately. :shakes:Quote:
Originally Posted by EniGmA1987
Don't know for sure. As I understand it, as long as they are all the same clock speed, there shouldn't be an issue, but it all comes down to what your vendor BIOS will support.Just because it is possible doesn't mean that they will be supporting it.Quote:
Originally Posted by jeanjean15
Folks, we are shipping Interlagos for revenue right now. If there was a problem we would not be shipping.Quote:
Originally Posted by AsterixXx
Ther overhead for sharing is pretty low. You'll most likely get better results with running on 2 modules, power gating down the other two and taking the higher boost state instead of trying to run one thread per module. Your results might vary, but I would be willing to bet that taking the boost upside is better in the long run.Quote:
Originally Posted by spicypixel
any more rumors of watercooling setup for AMD bulldozer?
or is Amd gonna get one up'd again by Intel?
http://www.legitreviews.com/article/1707/
_Wasn't he the team leader who lead RV770 project?Quote:
Originally Posted by Canis-X
Like when you shipped Phenom 9600 and Phenom 9500? And is that the reason for the delay from Q2 to Q4?Quote:
Originally Posted by JF-AMD
If they're not going to bundle it with any CPU's, why even bother? :)Quote:
Originally Posted by tbone8ty
Nothing points that way. Llano which is harder to produce isn't affected at all. BD on the other side has been delayed roughly the time it takes to do a respin, and has changed stepping number, it looks like there was the design itself that needed tweaking.Quote:
Originally Posted by freeloader
Ah yes, I forgot about the stepping change. Would that change have anything to do with Bergman leaving AMD though? PCMag is reporting that he took a job with an unnamed company.Quote:
Originally Posted by -Boris-
http://www.pcmag.com/article2/0,2817...id=dJWbeUOIc7f
There is another sisoft entry for "non-existent" Opteron ES 16C @ 2.6Ghz. This time results are even better than the last time when the model was listed as running at 2334Mhz. Now it is listed at 2.6Ghz and scores are almost proportionally higher.
Details for Device 2x AMD Opteron 6282 SE
Processor Arithmetic Benchmark 2x AMD Opteron 6282 SE @ 2.6Ghz 243.275GOPS (previous score @ 2.34Ghz was 231.9Gops )
Processor Multi-Media Benchmark 2x AMD Opteron 6282 SE @2.6Ghz 628.234Mpix/s (previous score @ 2.34Ghz was 585.9Mpix/s)
It appears that Turbo is not functional and newest integer throughput results for 4x Opteron 6272 @ 2.1Ghz confirm this ( it scores 377.7 GOPs with 2x more cores => 377.7/2=188.8Gops for 32 cores @2.1Ghz clock? => 188.8 x 2.6/2.1Ghz= 233.8Gops or virtually the same as "non-existent" 2x Opteron 6282SE @ 2.6Ghz ).
For comparison with MC 2P system @ 2.5Ghz :
Processor Arithmetic Benchmark 2x AMD Opteron 6180 SE @2.5Ghz 202.098GOPS
Processor Multi-Media Benchmark 2x AMD Opteron 6180 SE @2.5Ghz 331.207Mpix/s
Note that Turbo appears to be not working for integer throughput test in the case of 6282SE @ 2.6Ghz. Still it is faster than 2.5Ghz MC system by a good 243/202=1.2 or 20% in Processor arithmetic test and 628/332=1.89 or 90% faster in Multimedia test. Multimedia test uses AVX in case of new Opteron which gives around 11% better results on Bulldozer versus SSE test. So 1.89/1.11=1.7 or 70% faster than MC in legacy SIMD (SSE) throughput.
Looking at these latest numbers for 6282SE (which is supposed not to exist according to AMD :) ), FX 8150 @default with Turbo should be in the ballpark of : ~86-90GOPs and 205-210Mpix/s(184-189Mpix/s if you don't use AVX but SSE), in the arithmetic and MM tests, respectively. For comparison 1100T @ def. gets ~66.6 GOPs and 110-115 Mpix/s.
edit: note that these are still "ES" results and not from a retail/final platform
http://tof.canardpc.com/view/c481954...127ce1d7f0.jpg
http://tof.canardpc.com/view/5c93cf9...6a353dbae1.jpg
You will find the original source by yourselves ...
Shouldn't an hypothetical Phenom II X8 @ 4.8Ghz score between 10 a 10.5 points in CB11.5?
If it is sold seperatly then the warranty/RMA can be limited,... If the cooler leaks they'll replace it but any damage caused by the leak will not be covered. ;)Quote:
Originally Posted by wez
Yes I'm aware of those. I am speaking about Sisoft sandra opteron and Zambezi results,not C11.5.Quote:
Originally Posted by Olivon
the old architecture gets about .3 pts per ghz per core, round down a little for inefficienciesQuote:
Originally Posted by Andi64
so up to 8.4pts at 3.6ghz for an octo PII
11.3 pts for 4.8ghz
keep in mind that in the CB11.5 thread weve seen a half point spread even at the same ghz. but we also see people getting mid 7pts in mid 4ghz. so these results look bad
double post
Why you feel concerned ? :confused:Quote:
Originally Posted by informal
I just post informations that I just find at the moment, continue speaking about Sisoft Sandra if you want dude ...
No offense but that information you've "just" found is old information (and you know it). We have known this already for about a week or so. It was posted in this very topic. What I posted in my post was latest sisoft entry for Interlagos.
You're right informal, but that's the first time I see the visual about the thing and I share as I always did.
That slide deck looks like the biggest pile of malarkey I've ever seen.Quote:
Originally Posted by Olivon
Logos that don't align and have double strokes, vector arrows that blur out while text is still sharp, ridiculously huge text on slides with very little explanation.. missing footnotes, no benchmarks, and AMD hasn't used Cinebench screenshots for benchmarking to date (it's always a table + specs on the same slide)
Takes a Czech to create such dumbassery- and to the nationalities of whoever believes that... hah.
First time that I've see it as well.
haha the cinebench slide is really funny:
Attachment 120379
Now that you mention it , it's really odd that AMD will use intel hardware running OS X and put this into their own official slide deck. One would think they would use opteron or phenom or athlon for that screen capture of c11.5. Or even more logical is that they would use actual Zambezi with the actual score of 5.95 showing up in the screenshot. Strange stuff. This of course doesn't mean that the slide deck is "fake" or not real,just that it is odd.Quote:
Originally Posted by r.p
fake 100% :)
Not really fake, but means nothing.
I have a stronger feeling that naming scheme would be changed when Zambezi is released, just like obsolete 8130p, so we might needn't bother how 'FX-8150' perform.
No it's not. AMD used the screen from here: http://www.maxon.net/uploads/pics/ci..._screen_16.jpgQuote:
Originally Posted by Oese
IPC is lowered with Bulldozer.
slides are from oo..., the one who shall not be named. So we can just quit the descussion about those posts. Whether it is true or not, he isn't worth the effort.Quote:
Originally Posted by informal
Thats brings integer alot lower per clock than phenom.Quote:
Originally Posted by informal
http://www.sisoftware.eu/rank2011d/s...284f7cafa&l=en
Same core count, lower frequency and higher performance. So like you mentionned, BD seems to have alot of problems integer wise, fpu wise they are pretty good.
There's more fake to this then that ;) I believe AMD would not use some screen like that or make a presentation anywhere near to this. This is worse then 1st college class presentation :DQuote:
Originally Posted by rog
And yes I believe too IPC of a single BD "core" is maybe slightly lower then K10.5 core, but this doesnt mean these slides are not fake xD
Im just saying, even when some :banana::banana::banana::banana: is clearly fake, it still does not rule out that there is slide deck from such event.
Even when Interlagos shows up, it still will not tell us the performance of Zambezi. I don't know if I read it here or somewhere else that Interlagos is going to use an earlier stepping. The problem with the earlier stepping does not effect server workloads as much. Can anyone confirm this?
I don't believe that is true.
Quote:
Originally Posted by Andi64
You guys seem to have missed what chew said...Quote:
Originally Posted by Manicdan
1. Bulldozer functions like a 4 core that is able to execute 8 threads. You get roughly 5x scaling in CB from one to eight cores. (Better than HT) I don't care what PR tells you, that's how it works. (Unless new charts have come out with magical 7.x scaling?)
2. You have two options with the "STARS" core.
...Option 1. Shrink X6 (probably with redesigned Llano "STARS" core, so 3% IPC increase lets say) and increase frequency a tad. Lets say that this arch. will do 4.5 Ghz.
...Option 2. Add more cores, most likely decrease IPC from Thuban or keep it the same using Llano's tweaked core...lets say this arch will do 4.1 Ghz.
Bulldozer. More IPC than X4 in 4 threads or less, "less" distributed per "core" in 8.
Also, IPC is so far behind in Phenom II, why would we use STARS again? We are verging on 25-30% slower than Intel CPC, with a 15% frequency difference and new intel chips are doing 3.8 Ghz turbo, OCing to 4.8-5.0. We need a change. (If 5% IPC hit means 15-20% more clocks, I'm all for it tbh.)
You guys are so focused on multi-threaded results it's rediculous. Trying to predict single thread performance with a multithread benchmark is rediculous.
It's kind of like comparing a 2600K that does 5 Ghz on air to a 980X that does 4.3. Do you really want that 980X because its Cinebench score is higher, or would you like the extra single thread perf?
With no real CLEAR results WITH explanations out yet, I am still firm with my belief that in layman's terms, CMT = AMD HT in physical form that can not be turned off, trumping Intel's HT.
Any word on stable underclock potential yet? I'm referring to power savings, when compared to stock.
no one knows the the exact scaling because its hidden behind NDA.Quote:
Originally Posted by BeepBeep2
and if they did stick with PII architecture it would be more than 3% because llano does quite nicely for not having and L3. it would also pack in more true cores. by omitting the extra space used for the second thread of the module, the core itself got bigger with BD, so IPC should be higher or they found ways to waste more space. (btw your option 2 makes no sense, why reduce IPC since they can easily pack in 8 old cores in BDs size, and why would that drop the clocks by 400mhz)
i honestly dont believe in just 6pts for BD at 3.6ghz, and if that is the case its probably on the very low end of the spectrum for how it handles otherwise it would be a step backwards.