Core i7 920 rev D0 stepping 5 :)

Printable View

Show 100 post(s) from this thread on one page

03-04-2009, 04:32 AM
PcCI2iminal

Core i7 920 rev D0 stepping 5 :)

http://img134.imageshack.us/img134/1369/018tjz.jpg
http://img134.imageshack.us/img134/1521/024s.jpg
http://img3.imageshack.us/img3/6387/028nas.jpg
http://img3.imageshack.us/img3/2624/4150mhzok.jpg

Vcore @ spec -> 1.25v @ load and 1.05v @ idle
Ram @ 1.72v
VTT/PLL/NB @ minimum
cpu cooler - Thermalright Ultra 120 Extreme-1366
cpu fan - San Ace 120x120x38mm @ 102CFM
room temp @ 18c

Thanks Biostar for the new beta Bios with support for rev D0 ^^
03-04-2009, 04:47 AM
eva2000

ooooooooh lucky man indeed PC :D

max bclk will be interesting :)
03-04-2009, 04:50 AM
Viss

Nice :) looking forward to max bclk. Have fun mate :up:
03-04-2009, 05:01 AM
gurusan

Looks promising, any "stable" shots?
03-04-2009, 05:55 AM
bito

Looking to get one of these soon :)
03-04-2009, 06:25 AM
dpa

give us some vantage/06 cpu tests scores :D load the 4 cores :)
03-04-2009, 07:39 AM
PcCI2iminal

max bclock -> boot @ 222MHz and its prime stable

http://img19.imageshack.us/img19/857/primeok1.jpg

a half MHz more and the system will freeze :(
I need a new Bios :)
03-04-2009, 08:04 AM
PcCI2iminal

http://img15.imageshack.us/img15/149...hz155vcore.jpg

vcore @ 1.56v
03-04-2009, 08:10 AM
masxerofkartis

Hey there, is there a release date for these fellas yet? also do you have any idea on price margins

And ofcourse on who already has a 920, is it worth it to change it over this for 24h use?

Thanks
03-04-2009, 08:11 AM
Splave

nice so it clocks just like a c0 :)
03-04-2009, 08:23 AM
NBF

Very interesting toy you got there :D

Thanks for the heads up.
03-04-2009, 08:26 AM
zlojack

I'll wait to see what the retail versions do.

Too many times we see great screens of ES chips or cherry chips running 10 minutes of Prime at great clocks, but the actual retail chips don't come anywhere near them.

EDIT: Very nice 4.1 GHz on those volts though!
03-04-2009, 08:29 AM
FabricioGS

Congrats from Brazil PcCI2iminal :)

How about temps, equal to C0/C1?
03-04-2009, 08:40 AM
okorop

very nice test....what temperature do you have? do you have less heat than the c0-c1 or not?
03-04-2009, 08:41 AM
TheKarmakazi

Im eager to see if this stepping will actually improve anything...
03-04-2009, 08:43 AM
diablos

sbav, i wanna it
03-04-2009, 09:26 AM
Viss

Thanks PcCI2iminal :) Nice vcore for the clocks, if max bclk is same as on C0 there's not much to change i920 C0 for. Looking forward to i950 and 965 :up:
03-04-2009, 10:02 AM
Smoicol

good cpu,
there are improvements?
if you where them known?
temperatures? voltages?

regards
03-04-2009, 10:04 AM
okorop

Quote:

Originally Posted by Smoicol

good cpu,
there are improvements?
if you where them known?
temperatures? voltages?

regards

I agree with you we need some more information :)
03-04-2009, 10:06 AM
crazyea

First screenshot is single core @ 22 multi??? Second without HT? I would like to see results using all cores w/ht. And temps to boot!
03-04-2009, 10:13 AM
Splave

if your room temp is 18c, how is 19c possible at idle on air?
03-04-2009, 10:17 AM
okorop

Quote:

Originally Posted by Bobbylite

if your room temp is 18c, how is 19c possible at idle on air?

I think the temps are not shown correctly :)
03-04-2009, 11:28 AM
doomwill

congrats !!!!vlw mlk , grande novidade meus parabéns pelo exemplar.
03-04-2009, 11:32 AM
roller11

4641 Mhz CPU clock is good, but that is merely the 'boot into windows' clock, no load or stress testing. And of course we don't know what cooling was used...dry ice, LN2, phase? My C0 920 passes 1M superpi at 4662 CPU clock at just 1.484v Vcore, 4 cores, HT off, all on a Tuniq tower 120 air cooler.
03-04-2009, 02:37 PM
PcCI2iminal

Quote:

Originally Posted by Viper.Br

Congrats from Brazil PcCI2iminal :)

How about temps, equal to C0/C1?

Quote:

Originally Posted by okorop

very nice test....what temperature do you have? do you have less heat than the c0-c1 or not?

yes guys,the D0 runs a bit colder than C0 :)

Quote:

Originally Posted by loc.o

Thanks PcCI2iminal :) Nice vcore for the clocks, if max bclk is same as on C0 there's not much to change i920 C0 for. Looking forward to i950 and 965 :up:

Bart ,its very easy to get high clocks and i think the Tpower is holding me back on Bclock,i will talk with engineers

Quote:

Originally Posted by Smoicol

good cpu,
there are improvements?
if you where them known?
temperatures? voltages?
regards

depends ,some good i920 C0 can run with high clocks and low voltage and some not but will be nice if all these new D0 will run like that ES

Quote:

Originally Posted by Bobbylite

if your room temp is 18c, how is 19c possible at idle on air?

the sensor is off i think or the real temp is the higher showed on coretemp ->26c :)

Quote:

Originally Posted by roller11

4641 Mhz CPU clock is good, but that is merely the 'boot into windows' clock, no load or stress testing. My C0 920 passes 1M superpi at 4662 CPU clock at just 1.484v Vcore, 4 cores, HT off, all on a Tuniq tower 120 air cooler.

its all about luck ,Ive owened tons of good i7 920 C0 and just a few of them was good for 4600MHz on air ;)

Quote:

And of course we don't know what cooling was used...dry ice, LN2, phase?

its all air my friend ,LN2 go to Xtreme section ;)
03-04-2009, 02:48 PM
PcCI2iminal

http://img4.imageshack.us/img4/7933/imagem2g.jpg
http://img404.imageshack.us/img404/9715/3d064400mhz.jpg

air - room temp @ 19c
GTX285 @ 750/1500MHz
03-04-2009, 03:04 PM
FabricioGS

54ºC @ load :eek2:

Is this correct? You're using the True air cooler right?
03-04-2009, 05:18 PM
overcrash

Quote:

Originally Posted by Viper.Br

54ºC @ load :eek2:

Is this correct? You're using the True air cooler right?

I would like to know that as well :confused:
03-04-2009, 05:33 PM
okorop

can you post a screen with prime 95 open and core temp as well please :)
then the tj max was not 100 ° or with the d0 it changes?
03-04-2009, 10:58 PM
Astennu

Very nice OC ! My i7 920 C1 only does 3.8 on 1.35v bios. Mugen 2 cooler. for 4.0 i need 1.5v :S i realy whant a new D0.
03-05-2009, 03:06 AM
SeLecT

Nice results PcCI2iminal! :up:

Did you test the other D0? :D Can reach with the second CPU such results?
03-05-2009, 04:51 AM
psyxeon

We want some in Europe!
03-05-2009, 05:20 AM
fr0wn3r

:shocked:
Keep up the good work! :)
03-05-2009, 06:34 AM
Krautmaster

temps for post 26 pleaze ;) - a c0 can do this @ 1,52V too, but temp will be near to 100 degree @ prime load and air...
03-05-2009, 07:18 AM
MrHydes

looks good enough for me, but i'm quite unsure if it's good enough

to switch C0 to D0, would like to know what you think PcCI2iminal about it

cheers and thanks for the nice post
03-05-2009, 07:47 AM
Chri$ch

Quote:

Originally Posted by Krautmaster

temps for post 26 pleaze ;) - a c0 can do this @ 1,52V too, but temp will be near to 100 degree @ prime load and air...

look at the taskbar, 50-54-54-57°C load
03-05-2009, 08:06 AM
dejanh

Quote:

Originally Posted by Chri$ch

look at the taskbar, 50-54-54-57°C load

I very much doubt that is correct. A new stepping to have shaved off 40C+ off the chip and made one of the coolest running Intel CPUs ever? No way. I do not believe it, not even for a second.

The temps are wrong, guaranteed. An OCd 130W chip by design will never run <60C, ever (a 61.4% OC too mind you). :rofl:

Also, the 22x multi is unfair in one of the early screens. Does not show anything when a retail i7 920 does not have this multiplier at all. We don't even know if it is stable at this at all.

Finally, CPUZ voltages? No no, let's get actual measured voltages with a DMM. I'm sure you are at least using LLC.
03-05-2009, 08:16 AM
okorop

Quote:

Originally Posted by dejanh

I very much doubt that is correct. A new stepping to have shaved off 40C+ off the chip and made one of the coolest running Intel CPUs ever? No way. I do not believe it, not even for a second.

The temps are wrong, guaranteed. An OCd 130W chip by design will never run <60C, ever (a 61.4% OC too mind you). :rofl:

Also, the 22x multi is unfair in one of the early screens. Does not show anything when a retail i7 920 does not have this multiplier at all. We don't even know if it is stable at this at all.

Finally, CPUZ voltages? No no, let's get actual measured voltages with a DMM. I'm sure you are at least using LLC.

I agree with you :)
03-05-2009, 08:24 AM
dejanh

You'd think Intel would mention a change of this magnitude lol. No offense to the OP, but there really are some serious problems with the results.
03-05-2009, 08:34 AM
Hens

well tjmax should be 100 instead of 92. even still that is considerably cooler...
03-05-2009, 08:36 AM
okorop

Quote:

Originally Posted by Hens

well tjmax should be 100 instead of 92. even still that is considerably cooler...

I agree with you so 8 degree more that weshould count :)
03-05-2009, 08:39 AM
dejanh

Quote:

Originally Posted by okorop

I agree with you so 8 degree more that weshould count :)

Nah man, that does not account for anything. The readout is completly wrong. Even if you added the 8C that would still leave the chip running at mid 60s on air at 4.305GHz. A 130W chip overclocked by 61.4%! C'mon, think about this really...
03-05-2009, 08:43 AM
okorop

Quote:

Originally Posted by dejanh

Nah man, that does not account for anything. The readout is completly wrong. Even if you added the 8C that would still leave the chip running at mid 60s on air at 4.305GHz. A 130W chip overclocked by 61.4%! C'mon, think about this really...

of course you are right those cpu's work like the previous ones....the temps are totally wrong and we don't recive any answer from him about this problem
03-05-2009, 08:48 AM
roller11

When and where are these D0 chips for sale, and at what cost? Also, are they a different sku? Anything different (like a spec/FPO num) on the retail box so a buyer would know he's getting the D0 before he opens the box?
03-05-2009, 08:53 AM
dejanh

Quote:

Originally Posted by roller11

When and where are these D0 chips for sale, and at what cost? Also, are they a different sku? Anything different (like a spec/FPO num) on the retail box so a buyer would know he's getting the D0 before he opens the box?

Well, you can assume that all chips that start with 39 will be D0 because they would have been produced in 2009. Also, potentially some late week production 38 chips. I also think my boxes at least indicated C0 on the box sticker, but I don't feel like digging through my stuff to check now. They will have the same tray price as current chips (I checked this a while back with Intel).
03-05-2009, 08:54 AM
chuxle

I agree. PcCI2 post and information are flawed. There's not much to derive from his test. We will have to wait for the "real" D0 Retail version to come out and have it tested and tortured from a diversified group of XtremeSystem members before we can get an accurate gauge of this D0 version.
03-05-2009, 09:27 AM
ewitte

Those volts are crazy lol. I have not went over 1.35v and that was good for 4295 pi and only limited by QPI. Probably good for 4.4Ghz at 1.35v if I rely on turbo ;)
03-05-2009, 09:28 AM
okorop

Quote:

Originally Posted by ewitte

Those volts are crazy lol. I have not went over 1.35v and that was good for 4295 pi and only limited by QPI. Probably good for 4.4Ghz at 1.35v if I rely on turbo ;)

I agree :)
03-05-2009, 09:49 AM
bito

The D0 steppings will be labelled SLBEJ as opposed to the current SLBCH :)
03-05-2009, 10:05 AM
dejanh

Quote:

Originally Posted by dejanh

I very much doubt that is correct. A new stepping to have shaved off 40C+ off the chip and made one of the coolest running Intel CPUs ever? No way. I do not believe it, not even for a second.

The temps are wrong, guaranteed. An OCd 130W chip by design will never run <60C, ever (a 61.4% OC too mind you). :rofl:

Also, the 22x multi is unfair in one of the early screens. Does not show anything when a retail i7 920 does not have this multiplier at all. We don't even know if it is stable at this at all.

Finally, CPUZ voltages? No no, let's get actual measured voltages with a DMM. I'm sure you are at least using LLC.

I want to correct my statement on the 22x multiplier. As much as I cannot get the same deal on any of the 940s I owned regardless of the board, I do realize that you can get 22x multipler when only one core is active and turbo boost is enabled. The only way that I can ever see 24x multi is if I phsycally disable all but one core. Either way though, this does not change anything in terms of the rest of the results which are not realistic at all.
03-05-2009, 10:27 AM
MrMason

Very nice! I'm pleased to see that my C0 stepping can still compete, prime95 stable with 1.27V at 4Ghz

How did you get the 22x multi with all 4 cores and hyperthreading on?
03-05-2009, 10:37 AM
hlonipha

Quote:

Originally Posted by dejanh

Well, you can assume that all chips that start with 39 will be D0 because they would have been produced in 2009. Also, potentially some late week production 38 chips. I also think my boxes at least indicated C0 on the box sticker, but I don't feel like digging through my stuff to check now. They will have the same tray price as current chips (I checked this a while back with Intel).

I just talked with an intel tech support rep about an RMA and he told me that they did not have any 920 available for Advance Replacement until 17 March 2009. I wonder if the replacement will be a DO chip.. When are they supposed to be on shelves?
03-05-2009, 10:48 AM
Splave

Quote:

Originally Posted by MrMason

Very nice! I'm pleased to see that my C0 stepping can still compete, prime95 stable with 1.27V at 4Ghz

How did you get the 22x multi with all 4 cores and hyperthreading on?

Maybe because its an ES :shrug:
03-05-2009, 10:54 AM
Leeghoofd

Quote:

Originally Posted by ewitte

Those volts are crazy lol. I have not went over 1.35v and that was good for 4295 pi and only limited by QPI. Probably good for 4.4Ghz at 1.35v if I rely on turbo ;)

Not all CPU's are alike, mine even needed 1.4Vcore for just 4ghz stable ops, my new one seems better not even close to your cpu ( estimating 1.28-1.3Vcore for 4ghz ops ) It's not because it's a new stepping that it automatically means lower Vcore... If Intel just got their memory controllers up to scratch that would already be nice...

Quote:

Originally Posted by okorop

I agree :)

you agree on what ? too high voltage ? ow c'mon :shrug: he will not run them volts 24/7
03-05-2009, 10:58 AM
MrMason

Quote:

Originally Posted by Bobbylite

Maybe because its an ES :shrug:

Nah I've seen others do it, it's something in BIOS that needs to be changed.
03-05-2009, 11:06 AM
Leeghoofd

It's only on one core I believe, new Asus Biosses release it too... I920 : 22 multi, I940 : 23
03-05-2009, 11:11 AM
Splave

Ive seen it too but never with 4 cores 8 threads before
03-05-2009, 11:12 AM
MrMason

Quote:

Originally Posted by Bobbylite

Ive seen it too but never with 4 cores 8 threads before

http://img3.imageshack.us/img3/2624/4150mhzok.jpg
03-05-2009, 11:15 AM
Splave

^ Yeah I know, this is what im refering to :) Never seen anyone do that before on a retail chip
03-05-2009, 11:27 AM
roller11

ES chips typically have unlocked multipliers, we won't see 22x on retail chips.
03-05-2009, 11:31 AM
ReverendMaynard

Quote:

Originally Posted by roller11

ES chips typically have unlocked multipliers, we won't see 22x on retail chips.

absolutely, but cpuz also doesn't favor too well with unsupported steppings as well, so how can we be sure that anything in that screenshot is worth talking about?
03-05-2009, 11:34 AM
ewitte

Quote:

Originally Posted by Leeghoofd

Not all CPU's are alike, mine even needed 1.4Vcore for just 4ghz stable ops, my new one seems better not even close to your cpu ( estimating 1.28-1.3Vcore for 4ghz ops )

4.05 runs at 1.225v (only tested with HT on) though I only left prime open for 10 minutes because the temps where getting close to 100C. My guess is with the correct temps that would be stable because 1.2v still runs prime for about a minute. Until I have the temps under control not sure if it has to bump up to 1.25 or not.

Still going prime (7 hours) 3.6 @ 1.085v ;)
03-05-2009, 11:47 AM
dejanh

Quote:

Originally Posted by Bobbylite

^ Yeah I know, this is what im refering to :) Never seen anyone do that before on a retail chip

Quote:

Originally Posted by roller11

ES chips typically have unlocked multipliers, we won't see 22x on retail chips.

Quote:

Originally Posted by ReverendMaynard

absolutely, but cpuz also doesn't favor too well with unsupported steppings as well, so how can we be sure that anything in that screenshot is worth talking about?

I've been having that exact argument with people over on overclock.net in a thread about this thread. I have had 4 940s over time and I could never get the +2 multiplier to show up unless physically disable all but one core. ES chips do typically have unlocked multipliers so this is indicative of nothing.

Most of all, amen Reverend on this statement ;)

Quote:

Originally Posted by ReverendMaynard

This thread was started with the intent of giving false information.

I think that's the whole goal. To spread false information around and that's exactly what ticked me off a bit. If this thread were actually true, then Intel has released the single biggest lie upon its consumer base in the form of i7 C0. It would make the C0 a beta chip, an unfinished piece of technology that was sold to people as final retail product for what would ammount to field testing for the sole purpose of getting a leg up on AMDs PII or something. Even Intel does not go that low.

Again, I say, 57C on air at 4.305GHz, 1.52V :rofl: :down:

This thread should be removed.
03-05-2009, 12:09 PM
zlojack

Quote:

Originally Posted by dejanh

Again, I say, 57C on air at 4.305GHz, 1.52V :rofl: :down:

This thread should be removed.

LOL

:shakes:

Seriously.
03-05-2009, 12:17 PM
okorop

Quote:

Originally Posted by zlojack

LOL

:shakes:

Seriously.

I agree LOL it'is impossible with 1.5 volts having a core temperature of 59 degree
03-05-2009, 12:19 PM
ewitte

Quote:

Originally Posted by dejanh

Again, I say, 57C on air at 4.305GHz, 1.52V :rofl: :down:

This thread should be removed.

Thats like my idle temps. Even when the TDP is down to like 48W so possibly everything up to c0 just reads the temps wrong? Thats the only thing that could be going of if thats true.
03-05-2009, 12:29 PM
NBF

Its an ES with new stepping chip....

Take it easy people, those softwares can give wrong readings, temps may be off scale.

Lets give the man more time to fidle and report back as further testing is done.
03-05-2009, 12:30 PM
dejanh

Quote:

Originally Posted by NBF

Its an ES with new stepping chip....

Take it easy people, those softwares can give wrong readings, temps may be off scale.

Lets give the man more time to fidle and report back as further testing is done.

There is no time! Must know the truth! :p: j/k
03-05-2009, 12:31 PM
Fafeifa

If the D0 was that good, i'd give away my C0
03-05-2009, 12:34 PM
dejanh

Quote:

Originally Posted by Fafeifa

If the D0 was that good, i'd give away my C0

I think if it was Intel would be offering free replacements to people just to ensure nobody causes trouble over getting such bad chips that are C0 :p: Not to mention good luck to retailers trying to sell C0s, lol.
03-05-2009, 12:54 PM
CmB

Quote:

Originally Posted by Fafeifa

If the D0 was that good, i'd give away my C0

While your giving away CPUs, I'll gladly take them for DC projects. :D
03-05-2009, 12:57 PM
Fafeifa

Quote:

Originally Posted by mountain_king

While your giving away CPUs, I'll gladly take them for DC projects. :D

I think you've missed the sarcasm there :D
03-05-2009, 12:59 PM
CmB

Quote:

Originally Posted by Fafeifa

I think you've missed the sarcasm there :D

Actually, I didn't. Was worth a shot anyway, wasn't it?:D

--Chris
03-05-2009, 01:00 PM
NBF

Quote:

Originally Posted by ReverendMaynard

To post a screen shot and act like it's all kosher and that it is actually running at those speeds/volts is what sparks the imagination in riots of idiots who believe anything they read. Do you know how many forums are linking this thread and the kind of conversations that are going on about it?

This is not retail, and ES chips should never be shown to the public in the first place because indeed none of us are going to own a chip like that unless we obtain in illegally.

thread = fail....yank it.

I respect how you feel about it, but the OT shouldnt be blamed for what others imaginate from his screenshots IMHO.

All I see in his screens is that he has an ES D0 chip that runs 3dmark06 with 100MHz higher clock than my best C0 on air. :shrug:
03-05-2009, 02:02 PM
dejanh

1 Attachment(s)

Just for comparison sake I decided to do a quick max BCLK run on my 940 using the R2E board. For some reason even when I set the BCLK to 222MHz CPUZ still shows it at 220.8MHz, and TurboV (silly Asus tweaking tool) shows it as 221MHz. Either way, it's set at 222MHz x 15 and boots with 3.33GHz. I just randomly picked a voltage of 1.10625V (LLC is not enabled).

Should this be impressive (see screenshot)? Based on what everyone is making out the D0 to be it should...

The point is, this D0, there is nothing to it. It just creates a false impression that the chip is amazing. Realistically, there is nothing different we are seeing here than with a C0. Max BCLK/QPI clock is either restricted by the CPU or by X58 IOH. Not sure which one, and we won't really know until maybe the B-3 X58 revision is out to put this to the test with its IOH improvements. I believe Saaya had a good theory on QPI limitation being tied to the X58. Only a lower multiplier for QPI will at this point allow higher BCLK speeds and I think this was known about a week after i7 came out. I posted about it way back as well. Some boards do a bit better job, like the DFI, but most will top out (99.9%) at 220MHz-222MHz, or basically 4GHz QPI (8GT/s).
03-05-2009, 02:12 PM
PcCI2iminal

Quote:

Originally Posted by Bobbylite

^ Yeah I know, this is what im refering to :) Never seen anyone do that before on a retail chip

retail 920 @ 22x with all cores and if the HT is needed i can enable it ;)
only set Turbo enabled, EIST enabled and you have access to fixed 22x cpu multiplier
enjoy :)

http://img26.imageshack.us/img26/300...coresht22x.jpg

Quote:

Originally Posted by ReverendMaynard

This thread was started with the intent of giving false information.

i dont know who are you and you dont need to know who i am so do not judge things what you dont know.
i dont believe that temp showed by coretemp is the real thing, i just said the new D0 runs a bit colder than the C0
03-05-2009, 02:19 PM
Viss

Quote:

Originally Posted by ReverendMaynard

This thread was started with the intent of giving false information.

You should know better :down: Sad to see the way this thread goes... The 22x works with any reatail i920 on DFI as wel if you set C6 state.
03-05-2009, 02:26 PM
zlojack

What board gives the 22 multi on 4 cores?
03-05-2009, 02:35 PM
PcCI2iminal

Quote:

Originally Posted by dejanh

Should this be impressive (see screenshot)? Based on what everyone is making out the D0 to be it should...

The point is, this D0, there is nothing to it. It just creates a false impression that the chip is amazing. Realistically, there is nothing different we are seeing here than with a C0. Max BCLK/QPI clock is either restricted by the CPU or by X58 IOH. Not sure which one, and we won't really know until maybe the B-3 X58 revision is out to put this to the test with its IOH improvements. I believe Saaya had a good theory on QPI limitation being tied to the X58. Only a lower multiplier for QPI will at this point allow higher BCLK speeds and I think this was known about a week after i7 came out. I posted about it way back as well. Some boards do a bit better job, like the DFI, but most will top out (99.9%) at 220MHz-222MHz, or basically 4GHz QPI (8GT/s).

of course im not sure but i believe the max bclock is restricted by the mobo,bios maybe ,i never reached more than 222bclock with any cpu and i tried to push it hard with tons of Core i7(retail and ES) but I only used three good boards(Tpower ,RE2 and Gigabyte Extreme)
03-05-2009, 02:40 PM
PcCI2iminal

Quote:

Originally Posted by zlojack

What board gives the 22 multi on 4 cores?

Biostar ;),I havent tried with RE2 or Gigabyte but when the 22x is enable the bclcok will be limited to 205-207(air or ln2) and it goes unstable and only good for screen.
22x 180-190 range is fine for benching but i dont know if its prime stable :) but its better to use the same multiplier for all cores -> 20x :)
03-05-2009, 02:42 PM
dejanh

Quote:

Originally Posted by zlojack

What board gives the 22 multi on 4 cores?

Technically, all should if the rest of the cores are idle i.e. powered down due to C-state transitions. I have never managed to recreate this on the R2E though with 4 i7 940s unless I physically disable all but one core.

Quote:

Originally Posted by PcCI2iminal

of course im not sure but i believe the max bclock is restricted by the mobo,bios maybe ,i never reached more than 222bclock with any cpu and i tried to push it hard with tons of Core i7(retail and ES) but I only used three good boards(Tpower ,RE2 and Gigabyte Extreme)

I'm quite sure that it's the X58 chipset. My second bet is the CPU itself, but I find this a bit less likely now. Keep in mind that we are talking about many different motherboards, and many different chips, and 99.9% have the same BCLK limit, and the only thing that is consistently common between them is the X58 chipset.

I would not be surprised if the new B-3 revision of X58 allowed higher BCLK though we won't know until sometime after mid-May I think.

D0 is supposed to bring a few things to the table:

1) More consistent overclocks (no large variations in overclocking ability like with C0 chips where some can do only 3.8GHz and others can do 4.6GHz).
2) Slightly lower voltages for same clocks.
3) More balanced temperatures (related to voltages above).

Generally, D0 will not do much for improving the overclocking ability unless D0 was based on the best of the best of C0 in which case most D0 will clock better than an average C0.
03-05-2009, 03:02 PM
Splave

Quote:

Originally Posted by loc.o

You should know better :down: Sad to see the way this thread goes... The 22x works with any reatail i920 on DFI as wel if you set C6 state.

I think he is talking about everything but that :yepp:
03-05-2009, 03:05 PM
PcCI2iminal

Quote:

Originally Posted by dejanh

I would not be surprised if the new B-3 revision of X58 allowed higher BCLK though we won't know until sometime after mid-May I think.
D0 is supposed to bring a few things to the table:

1) More consistent overclocks (no large variations in overclocking ability like with C0 chips where some can do only 3.8GHz and others can do 4.6GHz).
2) Slightly lower voltages for same clocks.
3) More balanced temperatures (related to voltages above).

Generally, D0 will not do much for improving the overclocking ability unless D0 was based on the best of the best of C0 in which case most D0 will clock better than an average C0.

i think the Intel is working for improving the overclocking ability , things have changed since AMD reached more than 6Ghz with their cheap quad :)
03-05-2009, 03:09 PM
zlojack

Quote:

Originally Posted by dejanh

Technically, all should if the rest of the cores are idle i.e. powered down due to C-state transitions. I have never managed to recreate this on the R2E though with 4 i7 940s unless I physically disable all but one core.

The same is true for my EVGA X58 and my former GA-X58-UD5. If I disable three cores in bios, I get 22 with Turbo on.

I just thought they'd figured out a way to trick it into 22x multithreaded which is why I asked but I see that's not the case.
03-05-2009, 03:12 PM
Viss

Quote:

Originally Posted by Bobbylite

I think he is talking about everything but that :yepp:

So am i, should not have added the 22x comment :D
03-05-2009, 03:13 PM
PcCI2iminal

Quote:

Originally Posted by zlojack

The same is true for my EVGA X58 and my former GA-X58-UD5. If I disable three cores in bios, I get 22 with Turbo on.

I just thought they'd figured out a way to trick it into 22x multithreaded which is why I asked but I see that's not the case.

but why you need multi threaded 22x?22x is only good for single thread ,the others 3 cores will work at 20x and you need lots of vcore just to get the 22x core stable with high bclock :shrug:
03-05-2009, 03:22 PM
okorop

Quote:

Originally Posted by ReverendMaynard

yeah man, but he's using a Biostar. All of that info in the first post would have been significant no? Sorry man, but this shouyld be in the Extreme News section....not the regular, retail, blue collar section where people have to work together to find a way to decipher all of the bs that comes with this hobby.

I agree with you :)
03-05-2009, 03:29 PM
Viss

Quote:

Originally Posted by ReverendMaynard

yeah man, but he's using a Biostar. All of that info in the first post would have been significant no? Sorry man, but this shouyld be in the Extreme News section....not the regular, retail, blue collar section where people have to work together to find a way to decipher all of the bs that comes with this hobby.

I should not have said anything about the 22x multi or DFI.

Me point is that you should know the TS is a respected member that brought a lot of good info to all of us and never had or wil have the intention to give false information.
03-05-2009, 03:40 PM
ReverendMaynard

Quote:

Originally Posted by loc.o

I should not have said anything about the 22x multi or DFI.

Me point is that you should know the the TS is a respected member that brought a lot of good info to all of us and never had or wil have the intention to give false information.

All we need is pure details, not just a screenshot that basically tells everyone "Hey look what I can do". I meant no disrespect loco; I shoot from the hip, and sometimes it stings but my intentions are well meant.
03-05-2009, 03:43 PM
Viss

:up:
03-05-2009, 03:47 PM
chuxle

This thread needs to be closed.
03-05-2009, 03:51 PM
chuxle

I agree

Quote:

Originally Posted by ReverendMaynard

This thread was started with the intent of giving false information.

I agree :up:
03-05-2009, 03:52 PM
PcCI2iminal

Quote:

Originally Posted by ReverendMaynard

yeah man, but he's using a Biostar. All of that info in the first post would have been significant no? Sorry man, but this shouyld be in the Extreme News section....not the regular, retail, blue collar section where people have to work together to find a way to decipher all of the bs that comes with this hobby.

it was easy to reached the max bclock and the max clock but what i showed its not enough to post in Xtreme section, not since the C0 reached the same clock with same voltage or lower and some really good chips 4.8GHz or more.
and i can not post on news section,not when im not sure about everything,the sensor is off for sure and the chip is really good for air,thats what i know ;)

Quote:

And PcCriminal, I'm not trying to lamb-baste you, or call you out on bs, I'm just reacting to the crap loads of posts in other forums news sections with hundreds of noobs arguing over a chip that will not, in the state that you have, be available to the public.

im fine man and i dont want to start a fight ;)
i dont have much time to test and everything what i know about the temp i posted here(post #25)

Quote:

the sensor is off i think or the real temp is the higher showed on coretemp ->26c

Quote:

depends ,some good i920 C0 can run with high clocks and low voltage and some not but will be nice if all these new D0 will run like that ES

http://www.xtremesystems.org/forums/...3&postcount=25

and the new rev D0 will hit the streets at end of march and who wants to upgrade to lga1366 is better to wait and buy the fresh rev D0,of course if it will be not overpriced
and who wants to upgrade from C0 to D0 is better to wait and see what the new retail D0 can really do ;)
03-05-2009, 04:01 PM
runmc

I say let's have some respect for PcCI2iminal. No more thread crapping please. :)
03-05-2009, 04:34 PM
zlojack

Quote:

Originally Posted by PcCI2iminal

but why you need multi threaded 22x?22x is only good for single thread ,the others 3 cores will work at 20x and you need lots of vcore just to get the 22x core stable with high bclock :shrug:

Well, if I could have a 22 multi with a multithreaded workload on an i7 920 I'd take it :D

But I know it's not possible. What I'm confused about is, I was under the impression that you get 22 ,(and effectively) 0,0,0 or 21,21,21,21 with turbo on.

At least, on the two boards I've owned, the only way to see 22 is to turn off 3 cores.

Maybe I'm not understanding it properly. I've always said I have just enough knowledge to be dangerous, but not quite enough to know what I'm doing :p:
03-05-2009, 05:33 PM
dejanh

Quote:

Originally Posted by PcCI2iminal

i think the Intel is working for improving the overclocking ability , things have changed since AMD reached more than 6Ghz with their cheap quad :)

What this implies then is that in order to regain its throne as the king of OC Intel will blantantly EOL their 940 and 965 chips to replace them with much better i7 chips based off of D0 stepping. If they do that, I will make them replace my chip if it's the last thing I do.

Not to even mention that would be the biggest show of disrespect for the consumer I have ever seen.

I somehow find this outcome very hard to believe. I would have no issue with them keeping the specs and names on the 940/965 and improving the OC ability slightly compared to C0, but a new chip with new upper multiplier limit is just bull.
03-05-2009, 05:41 PM
SAE

I am eager to see more D0 results. Thanks for keeping us up2date, PCC. :) :up:
03-05-2009, 05:45 PM
Zucker2k

Quote:

Originally Posted by dejanh

What this implies then is that in order to regain its throne as the king of OC Intel will blantantly EOL their 940 and 965 chips to replace them with much better i7 chips based off of D0 stepping. If they do that, I will make them replace my chip if it's the last thing I do.

Not to even mention that would be the biggest show of disrespect for the consumer I have ever seen.

I somehow find this outcome very hard to believe. I would have no issue with them keeping the specs and names on the 940/965 and improving the OC ability slightly compared to C0, but a new chip with new upper multiplier limit is just bull.

Dude, how long have you been in this habit of overclocking? You're hilarious. Oh, you owe PcCI2Iminal an apology for crapping his thread. My God, since when did a guy have to apologize for showing pre-released hardware on XS. Y'all need to take a chill pill and be more appreciative of the fact that you belong in a forum where members get their hands on the new toys far more consistently than other forums.
03-05-2009, 06:16 PM
FabricioGS

Quote:

Originally Posted by Zucker2k

Dude, how long have you been in this habit of overclocking? You're hilarious. Oh, you owe PcCI2Iminal an apology for crapping his thread. My God, since when did a guy have to apologize for showing pre-released hardware on XS. Y'all need to take a chill pill and be more appreciative of the fact that you belong in a forum where members get their hands on the new toys far more consistently than other forums.

Perfect :up:
03-05-2009, 07:15 PM
freaksavior

WOW! if the retails oc like this then it may be worth swapping out the i7 for the revision

Show 100 post(s) from this thread on one page

All times are GMT -8. The time now is 01:04 AM.

XtremeSystems