'Bloomfield' is the CPU codename just like Yorkfield, Kentsfield, Wolfdale, etc.
Nehalem refers to the CPU architecture family like Penryn does to 45nm Core 2s.
Printable View
'Bloomfield' is the CPU codename just like Yorkfield, Kentsfield, Wolfdale, etc.
Nehalem refers to the CPU architecture family like Penryn does to 45nm Core 2s.
Isn't Penryn the 45 nm version of Merom ?Quote:
Originally Posted by bowman
AFAIK, the name of a micro-arch keeps being the same whatever the process is. And in case of Core 2, uarch is "Core 2", whatever the process / platform.
There are differences in the micro-arch of 45nm and 65nm Core2. Differences like those between Katmai, Coppermine and Tualatin/Willamette, Northwood and Prescott. Those generations of CPUs were branded as Pentium 3/Pentium4 CPUs but their architectures, although very similar, were different.
Conroe and Penryn have also very similar m-arch and are belonging to the same m-arch generation: Core2.
Conroe is a reference for the 65nm Core2 m-arch. Penryn is a reference for the 45nm Core2 m-arch.
At same time Conroe is also a codename for the desktop 65nm Core2 Duo E6xxx with 4MB/2MB(where 2MB are disabled out of total 4MB) L2, while Penryn is the codename for the mobile 45nm Core2 Duo CPUs.
Bloomfiled is a codename for the desktop 45nm Nehalem quadcore CPU, analogically like Yorkfield/Kentsfield is the codename for the desktop 45nm/65nm Core2 quadcore CPU
Yeah, but you're not going to call a 45nm version of it Merom or Conroe, you're going to call it Penryn. Same with the 32nm version of Nehalem, it'll be referred to as Westmere just like Intel does to avoid confusion.Quote:
Originally Posted by cpuz
65nm Core 2's were initially referred to as Merom (mobile version), Conroe came later.
Yeap, there's ...Quote:
Originally Posted by cpuz
Quote:
Originally Posted by savantu
1 stick of 2G DDR3-1066 ...
Can you read what is written on it ?
If we can program the PLL, we can try to o/c (I guess the BIOS has no option at all for that, does it ?)
Nope ... while Intel has unified their architecture across the 3 markets, they still give separate code names for each product. Merom, Conroe, and Woodcrest (Clovertown) was the 65 nm for mobile, desktop, and server (server quad) repectively... Penryn, Wolfdale, and Harpertown were the 45 nm tick version. Nehalem is not clear to me, perhaps server in this case, Bloomfield is the high end DT.Quote:
Originally Posted by Anemone
Jack
Clovertown/Tigerton -> Beckton
Harpertown -> Gainestown
Yorkfield -> Bloomfield
Kentsfield/low price Yorkfield -> Lynnfield
Allendale/Wolfdale -> Havendale
That's correct.Quote:
Originally Posted by bowman
The Nehalem code name is unique in that it applies to the overall microarchitecture that includes all the above products. Correct me if I'm wrong, but this is the first time in recent history where Intel has used a codename for a microarchitecture that did NOT also apply to a specific product.
Here's the comprehensive list of Nehalem product code-names...
http://members.shaw.ca/virtualrain/n...alem-chart.png
I'm not that familiar with Rightmark but it looks like there's something wrong with Rightmark's Max bandwidth then... it reports 12.8GB/s but a single stick of DDR3-1066 has peak theoretical bandwidth of 8.5GB/s.Quote:
Originally Posted by JCornell
The measured B/W of 5.6GB/s is actually not that great.
The R/W numbers could be influenced by cache.
Any chance of an Everest memory performance screen shot?
Yepp, 5.6GB looks like my 2.6GHz Northwood with DDR while it was overclocked to 3.36GHz IIRC. Hey if I didn't say it already, thank for the great info:up:Quote:
Originally Posted by virtualrain
It does apply to a specific product. As far as manufacturing is concerned, it is the codename for the product that eventually becomes either Bloomfield or Gainstown depending on how it's binned and eventually how it is treated prior to packaging. Some of them will have 1 quick path lane disabled and then become Bloomfield, others remain whole and become Gainstown.Quote:
Originally Posted by virtualrain
HMMM, would a bloomfield function in a Gainstown board if all that is different is that 1 quickpath channel is disabled or would the cpuid come into play as my guess is they have different CPUID's and the one for Bloomfield wouldn't be loaded into the Gainstown BIOS.Quote:
Originally Posted by Blauhung
Am I on the right track or way off?:D
We might have another BP6 on our hands..Quote:
Originally Posted by Movieman
Well maybe not in the same price range. :rofl:
Very helpful, thank you.Quote:
Originally Posted by virtualrain
Jack
Np... It's part of the Nehalem FAQ... http://www.nehalemnews.com/
:)
You reading this thread..? Nvidia isn't making a chipset for the Nehalem! *Dee dee dee*Quote:
Originally Posted by Anemone
With latest release CpuZ v1.46 :clap:
http://i245.photobucket.com/albums/g...alem/Fritz.jpg
===
Look how the Core Temp works with HT :)
http://i245.photobucket.com/albums/g...lem/Core12.jpg
...
Nice stuff JC... any chance of an Everest Memory Benchmark?
:D
Just like I already mentioned before, can't run everest / sisoft (HW/SW detection problem I guess) :shrug:Quote:
Originally Posted by virtualrain
Here's one for you to try..Quote:
Originally Posted by JCornell
Download BOINC, attach to a project, WCG will do..:rofl:
Then run the benchmark, it should do it automatically when installed for the first time
That will say a lot on what the cpu has for 'stones";)
This is for a windows 32 bits OS:
http://boinc.berkeley.edu/download.php
This is for a windows 64 bit OS:
http://boinc.berkeley.edu/download_all.php
Grab the one that is ver 5.10.45 if you have a 64 bit OS
Hey JC, when we can see some OC results? :).
Hm :), OC with default vcore? It is not dangerous with default ,-)...
maybe try Nero recode DVD, crysis bench, worldbench, winmedia encoder (set simillary as herehttp://techreport.com/articles.x/14606/8 etc :))
^Yes do some real life encoding tests so we see those to.
:)
Thanks for posting all this early stuff, man.
Quote:
Originally Posted by JCornell
What speed or is that the default speed?Quote:
Originally Posted by Donnie27
Yes. I don't understand how that works :confused:Quote:
Originally Posted by MuffinFlavored
default speed of what ?Quote:
Originally Posted by Donnie27
The DDR3 that I plugged is 1066 ...
Please do the BOINC test MM requested :)
Why this matter ? that's just false volt detect ...Quote:
Originally Posted by [XC] NetburstXE
How to judge with software that doesn't fully support(calibrate) for latest(final) hardware ? :shrug:
Actually vcore for 45nm CPUs is approx nominal 1.1v ... (of course this is for default speed)
I forgot to test after heard from Mr.MovieMan request ... I will pleased and as possible to fulfill some casual bench, nothin' more ... Hope all the bench will useful for some users ...Quote:
Originally Posted by anubis
Afterall, this is just for references ... and it's endless to compare all of it by my own ... ;)
I don't think that this is accurate.Quote:
Originally Posted by JCornell
The field Core Temp assumes for VID is actually used as FID (multiplier). So don't trust the early VID readings.
Hello JC,Quote:
Originally Posted by JCornell
Do you know which will be the last stepping? from what i see the actual mem controller is not fully working (i'm right? or is a mobo issue?).
I cant wait to see a fully working nehalem and see its overclock, also some real world gaming would be nice or some HD decoding/encoding.
Thank you.. It should take you maybe 10 mins max to download,install and run the benchmark.Quote:
Originally Posted by JCornell
The benchmark itself is app 60-90 seconds max.
You will see this (or like it) on your BOINC manager screen and just highlight and on the left hit "Copy selected messages" and then paste into a post.
6/29/2008 4:48:30 PM||Suspending computation - running CPU benchmarks
6/29/2008 4:49:01 PM||Benchmark results:
6/29/2008 4:49:01 PM|| Number of CPUs: 4
6/29/2008 4:49:01 PM|| 3558 floating point MIPS (Whetstone) per CPU
6/29/2008 4:49:01 PM|| 8289 integer MIPS (Dhrystone) per CPU
The above was on a Q6600 at 3600mhz with some items running in the background.
I would imagine the Nehalem to be above the 4000 mark on Whetstone.
The question is how much above 4000..:)
Again, my thanks for doing this.:up:
It's not a exact bench but rather will give a good idea of the Nehalems computational power.
The RAM of course, thank you!Quote:
Originally Posted by JCornell
Wait a minute :confused:Quote:
Originally Posted by virtualrain
NVIDIA doesn't have a license for the QPI connection, but could make a chipset for anything that doesn't have QPI. The lack of QPI info on the lower market solutions, does that mean they use the old FSB and not QPI, and could NVIDIA possibly make SLI chipsets for these if they wish to?
And even if NVIDIA doesn't have the QPI license, couldn't they just unlock the driver and support SLI on Intel's own chipset, if they wanted to?
Just speculation and sorry for off-topic.
QPI's main use is for Socket to Socket communication on this chart. No need to worry about it on Single socket systems. As for links between the I/O Hub and the Processor, Intel already uses a Serialized Point to Point interface called DMI. It is an offshoot of PCI-E.Quote:
Originally Posted by Seraphiel
nVidia also doesn't have a license for any of the new sockets and or IBEX (south bridge) as well according to at least on IBM platform guy I know. All Intel does is switch incoming I/O , to the processor instead of DMI, to Quick path and nVidia can only hope for Low end Legacy Products without an IMC:rolleyes:
The lower markets have just what amounts to an I/O southbridge. The northbridge is entirely integrated on the CPU, and PCI-E 2.0 connects directly onto the CPU. Even less wiggle room for NVIDIA. They could pull a Skulltrail and launch boards with a couple of NF200 chips to 'enable' SLI and and claim there's special sauce in those.. But that didn't fly so well the first time, I fear they'd get publically shredded if they pulled that stunt again.:rofl:Quote:
Originally Posted by Seraphiel
Of course they could, but that's exactly what Intel wants. SLI on Intel chipsets. At the same time NVIDIA's chipset business would go out of business, at least for desktops, because the 6 and 7 series for LGA775 sure has given them a frayed reputation.. Even if their Nehalem chipsets turned out 'okay', why would anyone bother if Intel had SLI and with the cloud of data corruption, crashes, bugs and bad drivers looming over nForce?Quote:
Originally Posted by Seraphiel
Thank you both for explaining this :)
You're totally under estimating the nVidia Faithful, Fanboys and Loyalists. They're afraid to even try anything but nVidia. Some even claim to have have switched from Intel chipsets and are much happier with nVidia. There's thread about Intel not giving nVidia a license and why. Amazingly nVidia Fans think Intel should Pay for SLI while nVidia should get all Intel for free! Giving more credence to the old saying, "Love Blinds". BTW, I'm very BIASed against nVidia but not without good reason.Quote:
Originally Posted by bowman
You're welcomed:)Quote:
Originally Posted by Seraphiel
If you want a bit more detail on how NVIDIA could support Nehalem with their own chipset, feel free to check this out...
http://www.nehalemnews.com/2008/05/e...halem-sli.html
It all depends on licensing of course and as you pointed out, a QPI license is only relevant on Bloomfield. The lower end parts don't use QPI.
Hell, all I want is a dual socket gainstown board with 2 quad core Nehalems in it running at 4000mhz and I don't care whose chipset is in it as long as it works!:rofl:Quote:
Originally Posted by virtualrain
well me too lol.
I was thinking about going Nehalem when it's cheap enough. I've also been facinated with the idea of two socket Nehalems. If Intel maintains their Xeon pricing for Nehalem (i.e lower clocked for pretty cheap) then i'll go dual socket when i get the money...or should i say if.
I still want to know how they plan on shoving all that cache on the 8 core Nehalem and still have it
a) yield
b) fit in a standard packaging.
1 I'd take NF200 chips to get SLI - just didn't have a fondness for FBDIMMS really. But SLI on Nehalem is something I'd take just not on a Nvidia chipset.
Now mind you, Nvidia chipsets could get better. Nvidia takes risks and gives little long term support to chipset buyers. It's a business they don't treat well, and that's really why they are in the pickle they are in. You look at Nvidia GPU drivers and you know that Nvidia takes bugs pretty seriously. Why? Because they know it's something people look for, and without that they would lose a chunk of business. With chipsets they have a more captive audience. And they abuse that which shouldn't be. If instead they had to test, retest, and issue patches in days to support chipsets, you can bet the chipsets would be vastly better built. Intel's current chipset business is only rosy (and even then not always) because of years of experience. Nvidia could do the same.
So giving SLI (which is perfectly fair as a trade for getting socket/connection rights to Intel's cpu's) isn't going to kill the chipset business. It's just going to grow it up fast, and make it play in the field like it should. Give people a solid, safe, and fast platform, with maybe a version or two aimed at clockers and enthusiasts.
It's simple really. And it's high time.
a) It will most likely be very expensiveQuote:
Originally Posted by Fuji
b) 4 socket server chips do not use "standard packaging"
Quote:
Originally Posted by Movieman
Far from your expectation :rofl:
http://i245.photobucket.com/albums/g...alem/BOINC.jpg
FYR
...
Poor Dave...hahaha :p:
Is it just me or the BOINC results are a little disappointing,being just on par with C2Q@65nm clock for clock?Or there is more than meets the eye?
That's not I want too :ROTF:Quote:
Originally Posted by Shintai
But, for real, some benches are not really much more differences if compares with Quad Kensfield / Yorkfield ... eg : 3DMark06 CPU Score, or ZD Win32 Floating Point, or CPUMark ...
Afterall, Just for references ... ;)
Thats quite normal for one new product to receive criticism on the first time before final release :D
I heard a rumor about a new HP + AMD for ServerWorks and for Intel LOL! This board looks like a Skulltrail type board for the folks like you. Since these cost more than some of the regular boards sold for servers, they have nothing to loose. Just a rumor though.Quote:
Originally Posted by Movieman
If this SW board doesn't see the light of day, you'll not have to worry about what Chipset, there'll only be one. IMHO, we're more than likely to see one by AMD than nVidia. Expensive low volume aren't treated the same as high volume cheaper stuff.
Maybe If I post the SS of 3DMark06 CPU Score, or ZD Win32 Floating Point Score, or CPUMark Score now, Could be more disappointing from any of you all :rofl:Quote:
Originally Posted by informal
But some benches are really showed the improvement of Nehalem compete with Kentsfield / Yorkfield ;)
Alot is also affected by the singlechannel memory and maybe bogus board. JC is your board also unable to run above 1 channel?
JC,you can post anything you like,after all this is just a ES with a not quite finished platform,so maybe things will improve a bit after it is polished.Post the mark06 and ZD Win32 FP Score, if you can of course :).
Thanks for all the input you provided,it's probably the first preview of the Nehalem system on the whole web :)
It's able to run 3 channel, Just simply update the BIOS ;)Quote:
Originally Posted by Shintai
You think every X58 mobo & every brands that showed at Computex are ready ? hehe ... Think again ... :ROTF:
How about some real world apps instead of so much synthetic stuff? Then how can we compensate for you only having one channel of RAM when these will ship with 3 Channels?Quote:
Originally Posted by JCornell
Example, Run Yorkfield or Kentsfield with one stick of RAM for comparison sake? Anyway, thank you very much!
Donnie 1 stick with Nehalem is a lot better than having 1 stick of memory with Kenstfield since C2Q needs all it can take from mem. bandwidth perspective while Nehalem has IMC.Don't forget this is DDR3 memory,a lot of bandwidth on disposal.I'd guess we'll for sure see some improvement with 2/3 channels of memory but this design choice was mostly made for servers and a future 4P beasts which will need it very much.Single socket/desktop users would probably see some benefits but not nearly as much as servers(look at AMD's NUMA for ex.).
I'd bet more on a things like early bios/board than mem. configurations (but the last will surely help it some)>
Hehe, nah. I know Anandtech also had troubles. One working board for a short time and the rest of teh rest on a faulty one :p:Quote:
Originally Posted by JCornell
So whats hold you back from the 3 channel? :D
Actually, I would like to compare Yorkfield vs Bloomfield vs Gainestown with compressed DVD data ...Quote:
Originally Posted by Donnie27
I just need time :zombie:
Nothin holding me :D ...Quote:
Originally Posted by Shintai
Just made me remembering last year, the first Wolfdale ES need up to 1minutes to complete SuperPI 1M :ROTF:
Aren't the score actually pretty good?Quote:
Originally Posted by JCornell
Kentsfield @ 3600 MHz (4 CPU)
3558 Whet / CPU
8289 Dhry / CPU
Bloomfield @ 2933 MHz (8 CPU)
2786 Whet / CPU
7095 Dhry / CPU
Now this is where I might get it all wrong:
(score) x (number of CPU)
Kentsfield:
3558 x 4 Whet = 14232
8289 x 4 Dhry = 33156
Bloomfield:
2786 x 8 Whet = 22288
7095 x 8 Dhry = 56760
Looks to me like Bloomfield / Nehalem is much stronger, or did I get it all wrong? If I have gotten the above wrong, then that WCG benchmark sure is misleading or confusing.
BOINC shows total scores(per CPU package),at least that's what i understand.Someone please confirm/deny this.
Now if this is the case,scale down the Kentsfield score to 2.93Ghz and you end up @ pretty much the same score with Nehalem,clock/clock.
I've looked at tomshardware.tw preview of Nehalem scores,and especially at Crysis CPU test.Managed to find a database of C2Q scores for the CPU test 2 at techarp:
http://www.techarp.com/showarticle.a...tno=499&pgno=3
Nehalem @2.93 scores:
while QX9650 gets around 18.7fps @1280x1024 noAA according to techarp database.Wolfdale e8400 even scores better than Qx9650with ~18.9fps in CPU test2.Quote:
Crysis: 1280x1024 noAA
CPU TEST2 18.29
Anyone knows what are the scores for QX9650 in CPU test 1 in Crysis bench tool @1280x1024?
I can confirm that the results in the benchmark are per core, not the total CPU or package performance. I get the same with 2 or 4 cores (within 100 pts). So the answer must be clear.Quote:
Originally Posted by informal
Also, Bloomfield have 4 "real" cores and 4 "HT" ones, or whatever they call it today. So the score must be an avarage of the stronger and weaker cores.
So I think, that if the HT cores were disabled, and only the real ones are left, then the score would actually rise quite a bit, but the number of cores would only be 4.
Just speculating, but I think Dave will actually be happy about the results :)
EDIT: The CPU tests in Crysis aren't very good ones.Quote:
Originally Posted by informal
From the very little infos about how the CPU tests were done, and with what settings beyond resolution and no AA, then I could have my Nehalem killer setup (all of these benchmarks and infos are true, but not telling the complete picture):
1280 x 1024 noAA
CPU TEST1 162.70 avg fps
CPU TEST2 94.29 avg fps
Point is, there is no sense in comparing the CPU tests with the limited info given by Tom's.
Hmm i think that JC was disappointed with BOINC since the scores where lower than Movieman expected.
Anyone else care to post their thoughts on BOINC benchmark results?
Cool:) Please take your time and get some rest!:up: Folks tend to screw up more when they're tired.Quote:
Originally Posted by JCornell
JC - Thanks for sharing all this good preview info!
Regarding the BOINC benchmark, it is per core, so for total work done the hyperthreading looks promising
If you assume linear scaling with clock speed (big assumption): 3600/2933 = 1.227x
Nehalem @ 3.6 would be:
Whetstone 3418
Drystone 8715
I think there might a difference in OS as well - 64-bit OS has advantage over 32-bit (I think Movieman's numbers were on 64-bit) - some guys are claiming 5-10% difference but maybe someone can confirm/refute this?
QFT! Folks just need to understand that these are Preliminary only, nothing else.Quote:
Originally Posted by Seraphiel
http://i245.photobucket.com/albums/g...oomfield_9.jpgQuote:
Originally Posted by informal
...
yes, im waiitng :). THX.Quote:
Originally Posted by JCornell
lol i like how the other program shows 16 logical cores. :ROTF:Quote:
Originally Posted by JCornell
It turns out its a real BOINC monster. We simply forgot SMT. Happy Dave then :p:
First of all Thank you very much for running the benchmark.Quote:
Originally Posted by JCornell
Now just so I am sure, this is a single socket quad core correct?
If that is so, then the new Hyperpath(Hyperthreading) is skewing your numbers. If it is possible to shut off that function so that only 4 threads are working you may see some VERY incredible numbers.
Not quite double but maybe 60-75% above what was shown.
If that is true, then you are sitting on essentially the most powerfull machine that exists outside of some supercomputer.
I don't say that lightly, on the old dual single core xeons the difference in benchmarks with HT turned on and off were app 70% greater.
Now add into this that your only on single channel memory on a board that will do tri channel, a pre release board and probably still some bugs and added all up my thought is Dear God Almighty!:rofl:
Oh, and I forgot to add: You only at 2933mhz..
At the risk of sounding like some schoolkid instead of a jaded 56 year old guy, stop and think what this will do when on a retail fully debugged board with tri channel DDR3-1333 or DDR3-1600 and running in the 3400-3600mhz range.
Just think on that a minute or two, before you call and buy Intel stock!
read above, your on the right track..:up:Quote:
Originally Posted by Seraphiel
Oh yea, quick, get the nitro pills for my aging heart!:ROTF:Quote:
Originally Posted by Shintai
Thanks Dave, and now off to sleep with you, dreaming about retail systems performance for a DP version (DP Beckton especially) :)
Since I have seen absolutely nothing as yet on Gainstown, the dual socket version of Bloomfield, I will have to dream on just single socket for now.Quote:
Originally Posted by Seraphiel
After a couple minutes to think on this I think that his single socket quadcore Bloomfield with working trichannel memory and maybe at 3400mhz would out work my just assemblend dual socket Harpertown at 3400mhz
Close but that reality is that nehalem may well be close to twice the computational power of an equally clocked Harpertown(Yorkfield)
That scary..
Ok, Now I need sleep.
Good night Gentlemen and thanks very much for posting the benchmark!:up:
Compare Fritz by 4phy & 4log cores with 1phy core
http://i245.photobucket.com/albums/g...alem/Fritz.jpg
===
http://i245.photobucket.com/albums/g...lem/Fritz2.jpg
...
Sure it's single socket, if not it wouldn't be Bloomfield ... :yepp:Quote:
Originally Posted by Movieman
Maybe tomorrow will end this Bloomfield benches, to Gainestown 3.xxGHz benches ... :clap:
...
OMG :shocked:Quote:
Originally Posted by JCornell
Please do a BOINC WCG run with that Gainestown benchmark, please!
Thank you, for all this insight and pre release benchmarks, it is really appreciated :bows:
Consider Done :rofl:Quote:
Originally Posted by Seraphiel
Yepp, thank you very much for all the info!Quote:
Originally Posted by JCornell
I just had to be sure so I asked. Agfain my Thanks for taking the time to do this.Quote:
Originally Posted by JCornell
As to Gainstown, are you trying to stop my aging heart?:rofl:
See if there is a setting in either Bllomfield or Gainstown to shut off the HT and then try and run that bench in BOINC again.
I think you will be very surprised, as will we all.;)
what u said there made me think.. if this debacle between Intel & NV is true (RE: chipset/SLi) - then one could essentially say that NV does not want to earn its keep - they rather get away with making crap chipsets (which they have been), than pull up their socks & put some effort into it :shakes: :down:Quote:
Originally Posted by bowman
if an individual/small business had this kind of work ethic they would quickly go bust! :yepp:
I wont be surprised, but just in a continued state of awe :yepp: ;)Quote:
Originally Posted by Movieman
Hi JC,
Still busy with the Bloomfield..or changed to Gainestown now.
Hope you like to do the chess tests for me!
Thanks,JP
You can always contact me on MSN.
My heart's all aflutter as well Dave. Dual socket 6 channels of memory for me.Quote:
Originally Posted by Movieman
Hehe, remember that Beckton is only MP platforms on the bigger socket.Quote:
Originally Posted by Seraphiel
Quote:
Originally Posted by JCornell
Not bad at all.. Faster than Q6600 @ 3600 MHz, and almost as fast as Quad Xeon 5320 8 cores x 2330 Mhz.
http://www.jens.tauchclub-krems.at/d...enchmarks.html
Just ran this on my dual harper at 3404 and got this but this is 8 cores and almost 500mhz faster.Quote:
Originally Posted by Face
:eek:Quote:
Originally Posted by Movieman
I'm going to have an anxiety attack waiting for the Nehalem.
X58 mobo
HD4850x2
6GB DDR3 (Tri-channel mode)
Oh my...! Think I'm going to need more Xanax. :shocked: :shocked:
And a pile of clean shorts to change into every hour reading this post.Quote:
Originally Posted by Xoulz
I gave up, just ordered in a truckload of Depends..:rofl:
Quote:
Originally Posted by Xoulz
Hope you got an extra Xanax ... this is scary :ROTF: ... Please don't suffer from panic attack :rofl:
http://i245.photobucket.com/albums/g...tom/Atom_1.jpg
...
:rofl:
I haven't seen any Atoms here. :(
They should at least make it possible to use Nvidia's chip, so that either individual mobo makers, or Intel themselves could offer SLI on the high end, much like Skulltrail, but say on X58 and Gainestown? SLI on Intel is mostly going to be taken up by the high end anyway, and that would leave Nvidia to sell high/med/low as they like, and still leave them the middle market where they probably make the most money.
Oh well. If quad 4870's turns out anything good in driver support and performance, SLI will probably die off anyway. Nvidia will still sell cards - I'm not that doom and gloom, but SLI itself could dwindle to a tiny niche market, which would mean it would lose regular driver support, which further perpetuates the dwindling.
I find it funny that they want ATI to support Physx for free, but they don't want Intel to do SLI. ATI could put all that work in and then get shut off when Physx 2 comes out? Oh well. Come 2009 when mainstream Nehalem hits, without a license, Nvidia will be effectively nearly out of the mobo business anyway. See what happens then.
But how does it perform when HT is enabled? ;)Quote:
Originally Posted by JCornell
It seems single threaded performance does not shape up as nicely as multi threaded performance.
We have seen SuperPi, cinebench, deep fritz and CPU mark 99 tests so far.
SuperPi shows impressive gains but isn't superPi quite bandwidth constrained so the reduced latency/increased bandwidth may help?
It looks like ~50% for 32m and ~25% for 1m.
Cinebench 1cpu (32bit) shows ~2% performance gains according to Anandtech, JC's tests look more like ~5%.
Single threaded fritz chess score is ~1,x% faster than penryn.
CPU mark 99 improved by ~9% over penryn.
Performance improvements ranging from 2-9% are not really bad, but neither are they breath taking. However, it may be not that important considering we are in the era of multi threading. I can't even come up with any mainstream single threaded benchmarks (besides games) I could ask you to test, so I could be sure about Nehalem's single threaded performance.
Just wondering whether you guys agree that we may see 0-10% performance gains on some compute intensive, single threaded, but not in any way bandwidth sensitive applications like those above?
Anyway it's not bad at all considering how well Nehalem will perform on average and that most of the transistor budget went into SMT/IMC benefitting only multi threaded and BW heavy applications.
Yes, I know its early alpha silicon, but I am quite sure there won't be much more of a boost than <5% from improved boards and chips.
Hehe, Yeap BIOS issue,HT disabled ...Quote:
Originally Posted by The Coolest
http://i245.photobucket.com/albums/g...tom/Atom_3.jpg
http://i245.photobucket.com/albums/g...tom/Atom_4.jpg
http://i245.photobucket.com/albums/g...tom/Atom_6.jpg
http://i245.photobucket.com/albums/g...tom/Atom_5.jpg
===
http://i245.photobucket.com/albums/g...tom/Atom_0.jpg
Without HSF burn in test :ROTF:
http://i245.photobucket.com/albums/g...tom/Atom_2.jpg
Onboard SSD
http://i245.photobucket.com/albums/g...tom/Atom_7.jpg
...
Quote:
Originally Posted by JCornell
:zombie:
Thats sick... honestly JC, I thought we were all friends here, then you go and pull that shiz! I really cannot function with so much of mind mind idling with vision of grandeur and an exotic build. Let alone unimaginable (achievable) benchmarks.
I have to go self medicate, l8tr.
@JCornell
Is it possible to run the new TrueCrypt 6.0 Benchmark function (100 mb buffer) with multi core support on the Nehalem system?
http://www.truecrypt.org/downloads.php
cu, BitpowerPM
I did too hehehe!Quote:
Originally Posted by Xoulz
im glad i didnt buy a new penryn cpu and just wait for nehalem lol. i only wish beckton wasent released in Q4 of 2009. but then again i can always go for dunnington.
http://img382.imageshack.us/img382/3...ngton01gm9.gif
image off of xtreview