Very limited numbers released but hey it's a start.
http://www.hexus.net/content/item.php?item=20291
Printable View
Very limited numbers released but hey it's a start.
http://www.hexus.net/content/item.php?item=20291
Thanks for the link. Clarkdale's IGP is doing surprisingly well.Quote:
Originally Posted by qurious63ss
performance is close to 785G IGP....
Yes very well indeed. I was under the impression that the Clarkdale IGP was nothing more than a gma 4500 integrated into the same package, but it looks like it may be something completely new or could it be that moving IGP closer to CPU make that big a difference?
I think there are three fundamental improvements at play here:Quote:
Originally Posted by qurious63ss
1) Slightly upgraded IGP arch (it's about due).
2) Increased IGP clockspeeds (I think there's a process shrink and there's definitely room for an increased thermal envelope from moving it into the CPU package [that has a much better guaranteed cooler] and out of the NB [that has a dinky passive thing plus all the other stuff on the same die])
3) Increased memory bandwidth due to improved memory controller on the Clarkdale (compared to G43/45's MC).
I could be totally wrong on all three points...I have no actual knowledge of the innerworkings of Clarkdale, if I did, I couldn't talk about it :p:
All three IPC, freq, and memory bandwidth would definately affect performance, question now is which one helped the most or was it a little of each. I agree with #1 but not sure if it's just a "slight" improvement, I guess that depends on how big a role #2 and #3 had on the improvements.Quote:
Originally Posted by Vapor
Indeed, looks pretty promising for Intel IGP.
This thing looks awesome, Big surprise on the igp performance and power consumpsion, Dual core smacking the quads around lol
The current G45 works at 800MHz while Clarkdale's IGP works at 700MHz so we can scratch #2.Quote:
Originally Posted by Vapor
Sweet...considering the large improvement in performance, that puts more weight into the other categories (i.e., my first and third suggestions + everything else actually at play). :p:
Hopefully drivers are getting improved as well. http://www.xtremesystems.org/forums/...ons/icon11.gif
Increadible performance of an Intel IGP. I'm really surprised! Now I hope the mobile parts won't get a scrapped version in order to keep down the power :/
I'm going with #3. I can't recall the Dr Who post, but he was singing the praises of how we'd all be surprised at how much the IGP would benefit from memory bandwidth and being on-die. Made sense at the time, and looks like these benches are backing it up.
Wow it seems that the igp has better drivers, those numbers seem better than the one posted in hkepc.
But as far as the cpu goes as a whole at around $143 "rumored" its not bad depends how much the h55 costs tough. On AMD's side i suspect the Athlon x3 and x4 to fight this foe. But price to performance is certainly with the Athlon x4 in the tests, a $99 cpu out performed a $143 cpu in most tests. I do believe the i5 dual cores will fair much better "Turbo" but at $176 base, come on really would you pay $30 more for turbo "When OCing i7 920 i have to shut turbo off".
Only the lowest end G9650 seems good in price to performance but intel has cut off its L3 cache, non the less at $87 it seems to be a bargain.
You can notice on the earlier reviews like HKEPC and PCPop that the same i3 540 got only ~8K in multi-threaded Cinebench. Now it manages to get 10K.
The IGP and drivers still have room to improve considering how far it is from launch. If you translate the HKEPC review it says that the graphics driver is one that gets the IGP "just working".
It's not really on-die, I think on-package suits it better xDQuote:
Originally Posted by Telperion
Yeah, Intel needs to charge less for their southbridge. Now they are charging as much for the P55 as the P45 used to cost. Ridiculous...Quote:
Originally Posted by ajaidev
Maybe NVIDIA will return :shrug:
There will be also Core i3-530 2.93 GHz $123.Quote:
Originally Posted by ajaidev
Also it looks like Clarkdale has an excellent overclocking potential.
4.0 GHz for only 0.832v.
Nvidia will have a problem with integrated GPU chipset with the S1156 Nehalem uArch. DMI being only 2GB/s and being shared with I/O is the biggest reason.
The big IGP surprise will be Nvidia I think, at least on AMD platforms. If the 880G only performs 15% better than current IGPs per rumor mill, then Intel has a chance to beat AMD in IGP performance(hah, think about THAT!).
Ya i know but the chip used by hexus is not a Core i3-530 but a Core i3-540 @ 3.06Ghz. The overclocking potential will also be very much there with Pentium G9650, the lack of HT may even help the OC figs, but 3MB L3 sux.Quote:
Originally Posted by kl0012
If Sandy bridge is an indication where things are heading after Clarkdale, you may need to upgrade your whole set up "Mobo+CPU" because the old mobo is not compatible with the new SoC. This lack of selective upgradeable sux but what to do its the future.
BTW AMD Fusion "LLano" is suppose to have a integrated gpu as powerful as Juniper "main stream"
i really need that bios update heheheh
@onethreehill Thanks :)
EDIT: Even tough Anandtech was a bit informative, all benchs are intel slides they dont represent real performance. I don't think a dual core nehalem can really out perform a Q9400, the comparison is fishy to say the least :(
I agree that most of the benches are synthetic and are not representing real-life.Quote:
Originally Posted by ajaidev
But I believe that Clarkdale at 3.33GHz will outperform a Q9400 in most of the software. It has 25% clock advantage, its core has higher IPC, it has HT, it has two ODMCs and it has turbo. So, it will not outperform Q9400 in heavilly(almost perfectly scalling) 4 threads optimised code where little memory bandwidth is required and memory accesses are rare. But it will be very close. In all other cases, ex. single-threaded and dual-threaded code, heavily bandiwdth dependent code, and partly optimised 3 and 4-threaded code, Clarkdale @3.33GHz will outperform a Q9400.
I think that the IGP was shrinken from 90nm all the way to 45nm actually. And honestly I am VERY surprised I never in my life thaught that Intels IGP could match AMDs offering.Quote:
Originally Posted by Vapor
EDIT: I suppose it was 45nm :p:
http://images.anandtech.com/reviews/...estmeredie.jpg
hexus link isn't working.
You are assuming that an "inferior" design will stay throughout the lifecycle of the architecture(with various modifications). With Clarkdale, the Intel will fix various weak spots which are pulling performance down in real world scenarios.Quote:
Originally Posted by LOE
Assuming 3DMark Vantage is correct, it'll be able to exceed the scores of Nvidia Geforce 9400 IGP per HKEPC:
http://global.hkepc.com/3878/page/6#view
G45: E3508
9400 IGP: E4445
1.5x G45: ~E5300
By the way Smartidiot89, the current G45 is already 65nm.Quote:
I think that the IGP was shrinken from 90nm all the way to 45nm actually. And honestly I am VERY surprised I never in my life thaught that Intels IGP could match AMDs offering.
v_rr: I think Hexus pulled the link probably by request of Intel
lol what, you guys got excited by intel docs about i3 igp performance? LOL...
come on guys, what do you trust more, some mfc doc claiming huge gains but being very vague about it, or some hw enthusiasts that actually benched and reviewed an actual i3 igp system?
doesnt anybody remember how intel initially claimed i3 igp perf will be ~2.5x faster than G45?
now its suddenly only ~50% more... yet you still trust those intel slideshows? :lol:
and this from anandtech:
OMG! 20% more shader cores :eek: :rolleyes:Quote:
While G45 had 10 shader cores, the 'dale GPU increases that to 12
about anandtechs preview... argh...
http://images.anandtech.com/reviews/.../partition.jpg
come on anand! you know better than that! i3 isnt 2 chip, its still 3 chip just that the igp/imc has been merged to one chip and is sitting next to the cpu silicon on the cpu package... why? to kill of ati and nvidia igp chipset by basically making them impossible as the only io to and from the cpu is a crippled 4x pciE 1.1 bus aka dmi...
its not 2 chips, its 2 packages at best...
http://images.anandtech.com/reviews/.../clarkdale.jpg
i3 drop in compatible with 1156 i7 800 and i5 700 series??? thats nothing but a lie from intel right there! :nono:
i3 needs a third pwm block for the GMCH, the igp/imc chip that sits next to the cpu chip on the cpu package...
far from every 1156 board has this, plus you need bios support too, i3 has a diferent imc than i5 and i7 and needs more than a basic bios update to work properly...
you want clarkdale i3 igp benchmarks?
how about some real world tests instead of performance predictions, estimates and bold claims from the manufacturer? :P
785G performance my 4ss... :D
courtesy of our friends in hong kong:
http://img17.imageshack.us/img17/1846/lolxr.png
this is a beta driver of course, even if a driver update within the next few months will kick this up and make it 250% faster than G45 accorss the board or even more than that... itll still be slow...
considering that this is still based on 965G and there has only been a 20% shader processor increase since G45, the perf boost is very nice...
but if you have sucky igp graphics and improve them by 250%... you still have sucky igp graphics... :shrug: :D
i donīt understand, why is that a lie?Quote:
Originally Posted by saaya
Clarkdale can run on every P55 Board. I tested it on 6 different Boards and had no Problems.
ironically your quoted benchmakrs shows 2-2.5x performance gain over G45 in realworld (games). :rofl:
Also it IS a 2 chips solution, there are only 2 physical "chips on" the motherboard (one chipset and one CPU). If you really would coun't every chip you would have 20x chip solutions... (audiocodec, NIC, bios, vrm controller etc, all this parts are a chip by definition)
Your constant nitpicking and complaining about everything realy gets old and annoying. :down:
Quote:
Originally Posted by Chri$ch
ouch right in the face.... :rofl:
The IGP is disabled on S1156, right?Quote:
Originally Posted by Chri$ch
Yes, IGP work only with H55, H57 and Q57 chipsets.
Dude.Quote:
Originally Posted by saaya
That Chinese benchmark is flawed that it uses a driver that barely recognizes Clarkdale. G45 gets way more than 3.4 fps at everything low on Crysis. From the same site it shows 8-9 fps. If Clarkdale can 2.5x that then it can at least equal 9400.
You can't compare the result to anything other than Intel IGPs because the driver even cripples the G45.
Look here: http://global.hkepc.com/1865/page/11#view
G45 Crysis: 8.6
x 2.6 = 22 fps better than 9400.
If you want to make yourself relevant you are better off not fooling others first.
Look again: http://www.anandtech.com/mobile/showdoc.aspx?i=3618&p=8
If they can manage 2-3x average in games it will be at least be on par with 785G.
It is not disabled on the socket. Just P55 chipset is missing FDI (display) interface. So Clarkdale will work on P55 boards but without IGP.Quote:
Originally Posted by FischOderAal
You are welcome :DQuote:
Originally Posted by ajaidev
Because it has been superseded by a new review?Quote:
Originally Posted by v_rr
http://www.hexus.net/content/item.php?item=20419
it can run on any 1156 board, but not with igp...Quote:
Originally Posted by Chri$ch
hmmmm so all 1156 boards HAVE to have the third pwm block for 32nm now?
i didnt see it on any 1156 highend board... then again they have so many phases its hard to tell :D
what boards did you test it on?
yeah but not 2.5x accross the board, in most cases its 2xQuote:
Originally Posted by Hornet331
:rolleyes: there arent 2 chips on the board, cpu and gmch are not one chip... they are on the same package but its not 1 chip...Quote:
Originally Posted by Hornet331
just skip my posts... :shrug:Quote:
Originally Posted by Hornet331
Thank you :)Quote:
Originally Posted by Chri$ch
Basically the same I meant ;)Quote:
Originally Posted by kl0012
i agree that its a bad driver, but its the best intel had at that time, 1 month ago... ^^Quote:
Originally Posted by IntelUser
yes, my bad for not mentioning it... they ran the same driver they used for clarkdales igp for the G45 to get the same image quality... thats why G45 has lower numbers in there than it usually would... if you compare the best G45 driver with the clarkdale graphics its more realistical but pretty sad, cause the 200-250% boost is almost completely gone...Quote:
Originally Posted by IntelUser
what do you base this on?Quote:
Originally Posted by IntelUser
here, have a look:
http://img4.imageshack.us/img4/3952/17634.png
if it would be 3x as fast as G45, it would manage to be the SAME as ion... not beat it... 790FX beats that, and 785 can be overclocked to 1ghz on air on most boards, which results in a really nice boost, sometimes above 30%...
even IF they reach a 2.5x boost, they will just about catch up with igp graphics from last year...
yes, using vthe clarkdale driver for the G45 system wasnt a smart move from hkpc... but if they would have used a propper G45 driver it would have made clarkdale graphics even worse, as theres barely a boost compared to G45...Quote:
Originally Posted by IntelUser
right... :lol:Quote:
Originally Posted by IntelUser
YOU might want to read that review your quoting, and i mean really read all of it and not as much as you think you need to prove me wrong :P
yes, but 790FX and ION both beat that...Quote:
Originally Posted by IntelUser
i think their overclocked clarkdale igp at 900mhz will probably do about as good as 9400 aka ion and 790FX... thats not bad... but its not good...
and we have yet to see a 2.5x performance boost...
to the smart4sses who said the hkpc numbers showed a 2.5x boost over G45, well read the review, they used the clarkdale igp driver on G45 which made it slower... normal G45 performance is only slightly slower than clarkdale igp numbers in that test... which makes sense seing as clarkdales igp only has 20% more shader processors than G45.
oh nice! very interesting!Quote:
Originally Posted by Vozer
sounds like they DID make quite some advancements with their drivers since last month!
EDIT: hmmmm
http://i279.photobucket.com/albums/k...md_785g_03.gif
http://img15.imageshack.us/img15/3184/17630.png
either G45 and possibly clarkdales igp lose half their perf when using 8xAF, or intel managed to double G45 performance within the last month?
either explanation sounds weird...
Vozer, did hexus say anything about image quality?
Would be quite right if you don't miss a litle detail... G45 is 65nm, clarkdale IGP is 45nm ... the shrink to 45nm alone save 40-45% space, . :p:Quote:
Originally Posted by saaya
Even with the IMC on the IGP core, theres more space for more transistors than can be used to increase the performance.
Who says the 12 alus got not beefed up. I doubt the 2 alus take up so much space.
Unfortunately, no.Quote:
Originally Posted by saaya
good point... im sure they did that... but unless they really fkced up and almost broke G45, how can they MORE than double performance by only reworking the shader processors and only adding 20% more? :confused:Quote:
Originally Posted by Hornet331
if they did it, hooray!
ion and 790fx performance is def nice for an igp system and good enough for an htpc and very basic gaming on a netbook or laptop... but i have some doubts about that... and even if the perf is there, G45 drivers are still far from perfect, and when did G45 come out? :D
i hope hexus or somebody else can post some more infos...
saaya, you keep saying 790fx, i think you mean GX?
huh, if the performance of the thing is near 9400 Nvidia is in trouble at least with people who get the i3. If Nvidia makes a ION mobo for the 1156 its bound to have a 9400/9300 igp and when fitted with a i3 you will get similar graphical performance that you would get with a integrated i3 igp.
So who would buy a ION for a i3??? Even if Nvidia does release a a 1156 ION mobo it may only make sense for lynnfields....!!!
The one thing about Intel IGPs are they can be competitive with other IGPs in one thing, but totally flunk in another. They are especially weak with games utilizing older shader models. Plus, they took a big hit with features like fog.
On Clarkdale, occulsion will be significantly enhanced with Hierarchial Z and it'll have more compute power with 3x the Mathbox units. Which might help a lot with the "weak" situations.
Not to mention they are extremely sensitive to bandwidth(something flawed with previous architectures?). Clarkdale will partly mitigate that.
The G45 driver did improve. They do much better with a recent article like this: http://www.tomshardware.com/reviews/...t,2381-11.html
how are these things for htpcs? Can they do 1080p or do they stutter?
theres an anandtech review on page one, it covers video playback
oh yeah, thx :DQuote:
Originally Posted by Manicdan
well ion came out when? theres ion2 coming afaik... :)Quote:
Originally Posted by ajaidev
but again, what do you think intel moved the igp chipset onto the cpu package for? theres no way for nvidia or ati to provide igp chipsets anymore, even IF the licensing thing would be no problem ;)
what does nvidia want to hook their gpu up to? dmi? LOL have fun!
yes, the g45 driver improved... but they dont say anything about image quality either... and as soon as you enable aa and af it seems to crash in l4d, and in wic it crashes as soon as you select medium details... they might want to get compatibility up before tweaking for perf :P
They can even decode 2 HD streams at once, where a 9400 just passes out. ;)Quote:
Originally Posted by Glow9