oO ... This makes more sense from a die layout... the 22 core chips are simply because of bad yields ...
There are no official 24c chip models.
ES Xeon Discussion
- Thread starter britinpdx
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
There are no official 24c chip models.
Thats incorrect. The next prozessors broadwell-EX will have full DIE 24 cores.
Intel just wants not to compete to broadwell-EP
It makes no sense that ES have full 24 cores and the final production not,
2 cores are just disabled.
bad yields are a problem on the new xeon PHI(knights landing)
EXAMPLE:
Windows Home VS Windows Professional
BAD YIELDS ?????? LOL
I don't think you understand the volume difference between ES and retail.
or the difference between EP and EX...
EX brought down all of the Itanium features to replace it.... it is a different beast all together.
EX is also a lower volume higher margin chip... even if they had a bit of yield issues, it wouldn't matter.
The Cinebench scores for that chip places it lower than the 22c chip... either they did something wrong with ram config or there is something wrong with the chip...
Software is not hardware... irrelevant example.
Fun trivia, AMD had a 20c chip that it never brought to production at the 6200/6300 gen...
maybe, just saw that as a statement in some intel press release.
"for the new Xeon E5 v4 only 22 cores are activated..."
cannot find that article yet...
cinebench i see that:
24 cores * 2300mhz = 55200(see HWinfo64 screenshot)
2726 cb
compared with
22 cores * 2400mhz = 52800
2678 cb
software is not hardware sure.
bad example.
its says "they could, but they dont"
"for the new Xeon E5 v4 only 22 cores are activated..."
cannot find that article yet...
cinebench i see that:
24 cores * 2300mhz = 55200(see HWinfo64 screenshot)
2726 cb
compared with
22 cores * 2400mhz = 52800
2678 cb
software is not hardware sure.
bad example.
its says "they could, but they dont"
Last edited:
have u ordered a fitting powersupply too ?
ES are mostly NOT compatible to intel server boards,
but u can ask each seller of ES for compatibility.
use supermicro server-boards or ASUS Z10 series.
Only ES with steppings R0/M0/B0 are running on gigaybyte x99 boards.
these are the official final steppings.
EVGA and ASRock x99 boards can run all ES steppings because the bios programmer want's this.
Last edited:
So my E5-2680 v4 have been running stable for a little more than 2 weeks.
I let it run misc. stress tests for 48 hours.
I installed HWInfo and compared instruction sets with Intels page - and they are identical.
Is there a stress test I can run that will use all features of the CPU and crash it if there is a bug or fault in the die?
I've installed ESXi. It has been running ~30 VMs for the last 12 days and is still 100% stable.
So I'm happy with my ES sample
I let it run misc. stress tests for 48 hours.
I installed HWInfo and compared instruction sets with Intels page - and they are identical.
Is there a stress test I can run that will use all features of the CPU and crash it if there is a bug or fault in the die?
I've installed ESXi. It has been running ~30 VMs for the last 12 days and is still 100% stable.
So I'm happy with my ES sample
CONGRATULATIONSSo my E5-2680 v4 have been running stable for a little more than 2 weeks.
I let it run misc. stress tests for 48 hours.
I installed HWInfo and compared instruction sets with Intels page - and they are identical.
Is there a stress test I can run that will use all features of the CPU and crash it if there is a bug or fault in the die?
I've installed ESXi. It has been running ~30 VMs for the last 12 days and is still 100% stable.
So I'm happy with my ES sample
Its hard to find a bug/fault
If you have one, repeat them.
Then find out its not the RAM, SOftware or other parts of your system.
Then compare your result with a Legal prozessor.
At the End think about a series fault like Haswells TSX bug.
Or remember the prime95 bug in Skylake that leaves the Intel labs,
it occours after 5 hours tress or after 3 weeks of normal operation too ? hwo knows ?
And those stress test are the reason why processors have no endless life
This is a earlier stepping, A0 = ES0, L0 = ES2
The broadwell A0 have better single turbo bins, L0 is near to pre-QS.
I think this stepping is more individual than L0(maybe no DP support)
TSX missing in A0 etc.
If you have the money it seems like a win win. Test it in a board they don't have listed and it works and you get $250 back if it doesn't they refund you.
I have one of these on my bench right now. I am waiting for the motherboard and ram to get here to put it together and try it. Parts should be here Tuesday.
Got it up and running but the board ended up having 2 bad usb ports so it is getting replaced.
odd things Processor marked as QG24, HWInfo says QG7V.
Cinebench R15 puts it at about the same as my dual 2670 system.
VT-D works fine, installed unraid, set up a VM passed video card and sound card and played a few games. WOW @ 1080P was running ~90fps at decent high settings.
odd things Processor marked as QG24, HWInfo says QG7V.
Cinebench R15 puts it at about the same as my dual 2670 system.
VT-D works fine, installed unraid, set up a VM passed video card and sound card and played a few games. WOW @ 1080P was running ~90fps at decent high settings.
Today my E5-2650v4 ES0 QH2N was delivered by FedEx(within 4 days)
icomputer_parts_international. The CPUs are fine, there were sold as ES so no misrepresentation. But it is really bad form to remark the case to show the retail markings when the CPU is ES.
Gotcha, thanks.
Will be ordering a few ES chips from bloommax soon, they seem to have a good reputation.
Except the remarking the E5-2630Lv4 ES are in best condition of all and have same turbo bins than the retail processors. I will order a second of it soon for replacement.
