Hot!eVGA SR-X (SR-3)

Page: < 12345.. > >> Showing page 3 of 7
Post
usmc0656
Superclocked Member
Re:eVGA SR-3 2011/12/29 08:32:24
Lehpron I will pay you to answer my FSB:DRAM question post!  I can't get a taker on it because it's closely tied to the CPU/RAM relationship through the MCH strap and such, but it doesn't seem many people understand how it works.  I beg ya !
pazza3169
New Member
Re:eVGA SR-3 2011/12/29 08:33:23
Higher stock clock speed of 3.1ghz mate. Nothing unlocked in terms of overclocking features. Usual turbo ratios available but thats about it. I have the latest C0 stepping.
 
Can any mod please comment on the overclocking or lack of it on this board as all xeon 2011 are locked and straps do not work.
pazza3169
New Member
Re:eVGA SR-3 2011/12/29 09:04:53
USMC what is you question??? I may be able to answer.
Brocasta
SSC Member
Re:eVGA SR-3 2011/12/29 09:28:24
I presume your chip(s) are still ES, pazza? I would hope that EVGA knows something we don't... perhaps some of the retail chips will have some of their shackles removed?
pazza3169
New Member
Re:eVGA SR-3 2011/12/29 09:37:57
Yes they are all ES. But unless EVGA has an ace in the pack no significant overclocking can be done on this board.
usmc0656
Superclocked Member
Re:eVGA SR-3 2011/12/29 10:27:10
The question was how to modify the FSB:DRAM ratio in the BIOS.  Nothing is straight forward.  I see the MCH Strap and the Unclock Frequency and the Memory Frequency, but all their settings are frequency numbers... not sure how to tweak those to what I want to set as my frequency settings so as a result I leave MCH Strap and Uncore Freq on defaults.
rottenmutt
Superclocked Member
Re:eVGA SR-3 2011/12/31 18:28:02
Forgotton
All motherboards and i do mean ALL can support ECC and or Registered memory. anyone who say's otherwise needs to go back to computers 101. So yes SR2 & SRX  can utilize it. The main difference between ECC and non-ECC is speed. NON Can be faster.......

this is true in the sense you can use a ecc dimm in a regular (non ecc) motherboard.  but the motherboard has been wired to read the ecc bits and has to use a memory controller (now integrated into cpu) which can utilize it (do the error checking).  error checking is not done on the dimm.  also servers have not been about pure speed, but more on accuracy and have been certified (gone thru more testing then desktop)
desktop boards will not have the ecc bits wired in most cases...
usmc0656
Superclocked Member
Re:eVGA SR-3 2012/01/01 08:16:00
Yep, proven reliabiity with registered ECC.  That's why I have over $150 invested per memory chip on my board... that is until my dominators get here.
mosie
Superclocked Member
Re:eVGA SR-3 2012/01/01 12:17:30
Im in way to test 8 piece of 2687W with 2 diferant steep ( 2 and 3).
 
If the steep BLK is locked its possilbe due to hight TPD problem ...   not just beacause ES verssion.
  possible all official rev C0 and C2 are steep locked.
 
2690 ??? possible unlock ?
 
If no XEON steep BLK unlocked    SRX usless.
  I can't belive that.
post edited by mosie - 2012/01/01 12:42:50
yaren
New Member
Re:eVGA SR-3 2012/01/02 17:43:34
dont worry it will be able to be done. just give it time... wait till all is released and then we make answers... there is no way they will deny this.
debs3759
New Member
Re:eVGA SR-3 2012/01/03 00:33:33
yaren

dont worry it will be able to be done. just give it time... wait till all is released and then we make answers... there is no way they will deny this.

 
That's my thoughts exactly. EVGA wouldn't be developing a board for overclocking chips that will never exist. In fact, I would be surprised if the developers don't have sample chips to test the overclocking features (if not then the board won't be ready for some time after the chips are released). Just because one person says they have a number of samples that they can't overclock on some other (sample) board does not mean that none will overclock on the SR-X :)
mosie
Superclocked Member
Re:eVGA SR-3 2012/01/03 11:47:44
You have not alreday buy 8 piece  2687W :)
 
  Its strange beacause the option is valid, but CPU fail to boot after modification steep seting.
 
Possible MB bios probleme ?
Brocasta
SSC Member
Re:eVGA SR-3 2012/01/03 12:01:33
debs3759

yaren

dont worry it will be able to be done. just give it time... wait till all is released and then we make answers... there is no way they will deny this.


That's my thoughts exactly. EVGA wouldn't be developing a board for overclocking chips that will never exist. In fact, I would be surprised if the developers don't have sample chips to test the overclocking features (if not then the board won't be ready for some time after the chips are released). Just because one person says they have a number of samples that they can't overclock on some other (sample) board does not mean that none will overclock on the SR-X :)

 
Devil's Advocate time.
 
This is a valid point, but the counterpoint is that EVGA's focus with the SR-X is to being CF/SLI to a Romley board, not necessarily for the overclocking functionality. The 8-core CPUs are going to overclock poorly, I'm betting, and those are what everyone who wants an SR-X is gunning for.
 
Personally, I think the ES chips are locked and (some of) the retail chips wont be.
 
 
mosie
Superclocked Member
Re:eVGA SR-3 2012/01/03 12:27:58
The steep are not locked I have acces to modifie this seting, but whene you aply this, MB fail to boot. the seting work but "certenely" MB fail to launch it.
 
Intel have not block this seting.
Personally, I think need just an update MB to fix the problem.
 
more info: http://hwbot.org/forum/sh...php?t=34292&page=2
Brocasta
SSC Member
Re:eVGA SR-3 2012/01/03 16:53:47
Oh, excellent. Good to know.
beaker7
Superclocked Member
Re:eVGA SR-3 2012/01/03 18:35:14
It's been nearly 2 weeks since i built a system.  Really looking forward to SR-X.
pazza3169
New Member
Re:eVGA SR-3 2012/01/04 14:59:49
The problem is the straps must be activated within the CPU. If they are not when you set 125 all otehr more sensitive buses also are clocked at 125 such as SATA and PCI EX and they simply wont run at these speeds. No divider is applied to them. I have experienced this with every ES CPU and every board I have tried.
mosie
Superclocked Member
Re:eVGA SR-3 2012/01/04 16:28:12
Yep.  Its certenely true.
I thinK same.
In fact: You can modifi directly the BLK, have the same efect.
 
But I dont know if this problem can be fix by MB bios, or if its a microcode CPU probleme.
 
 
post edited by mosie - 2012/01/04 16:34:37
BuffaloB
New Member
Re:eVGA SR-3 2012/01/04 18:49:38
debs3759

yaren

dont worry it will be able to be done. just give it time... wait till all is released and then we make answers... there is no way they will deny this.


That's my thoughts exactly. EVGA wouldn't be developing a board for overclocking chips that will never exist. In fact, I would be surprised if the developers don't have sample chips to test the overclocking features (if not then the board won't be ready for some time after the chips are released). Just because one person says they have a number of samples that they can't overclock on some other (sample) board does not mean that none will overclock on the SR-X :)

 
 
There are many more Xeons that have not been overclockable than one;s that have.  And Who knows how much overclocking headroom the new Xeons will have anyways, especially the 8 cores.
Forgotton
iCX Member
Re:eVGA SR-3 2012/01/05 00:24:00
socket 2011 server processors released by ibm can be seen at the links below. They are to PDF files 
 
http://www.intel.com/cont...11-datasheet-vol-1.pdf
 
http://www.intel.com/cont...ml?wapkw=socket%202011
debs3759
New Member
Re:eVGA SR-3 2012/01/05 08:45:45
Forgotton

socket 2011 server processors released by ibm can be seen at the links below. They are to PDF files 





 
They are the datasheets for SB-E i7, not server chips. Also they were not released by IBM, they were released by Intel (although IBM may have released a product that uses them - that I'm not sure about)
Forgotton
iCX Member
Re:eVGA SR-3 2012/01/05 08:51:25
They are fact sheets about the I7 2011 Socket server processors. All released data about them including all possible features.
 
Supporting Desktop Intel® Core™ i7-3960X Extreme Edition Processor for
the LGA-2011 Socket
Supporting Desktop Intel® Core™ i7-39xxK and i7-38xx Processor Series
for the LGA-2011 Socket
 
This means ALL of these processors are slotted to be usable by the SrX
And you can find everything the motherboard companies would be told these can do.
Since someone decided to dismiss without reading and i do not wish to flood posts.
http://www.intel.com/cont...processor-numbers.html

Processor Family Number Sequence System Type
Intel® Itanium® processor 9000 Multi-processor and dual-processor
Intel® Xeon® processor 7000 Multi-processor
Intel® Xeon® processor 5000 Dual-processor
Intel® Xeon® processor 3000 Single-processor  
even if one of the lowest in the fact sheet isn't.. other are and the abilities are listed this is more info then has been displayed about the processors that 2011 sockets then anywhere else in these forums. please research before responding. There have been so many hater posts even in build logs that ppl have responded to that after this any info found by me will no longer be posted save to those whom i know want it. msg me anyone if you wish to know from here on out. I found the xeon 2011 processor types with # of processors that will work because of information in the fact sheets. So YES if you read them you can figure out what processors are slated to be released for SR-X though the provided fact sheets.
post edited by Forgotton - 2012/01/05 09:44:32
beaker7
Superclocked Member
Re:eVGA SR-3 2012/01/05 08:54:41
Forgotton

They are fact sheets about the I7 2011 Socket server processors. All released data about them including all possible features.

Supporting Desktop Intel® Core™ i7-3960X Extreme Edition Processor for
the LGA-2011 Socket
Supporting Desktop Intel® Core™ i7-39xxK and i7-38xx Processor Series
for the LGA-2011 Socket

This means ALL of these processors are slotted to be usable by the SrX
And you can find everything the motherboard companies would be told these can do.


 
Nope.
kerrgreg
Superclocked Member
Re:eVGA SR-3 2012/01/05 09:28:42
I don't think we are 100% understanding you, otherwise you would not be writing that?  Surely you realise this is an SR-2 forum and I will be the first to admit I have not researched greatly the new 2011 xeon chips however I would be VERY surprised if the desktop processors support multi QPI links en order to run dual cpu's. 
 
So why should this evidence you have displayed prove that over clocking will be enabled on xeon lga 2011 chips
pazza3169
New Member
Re:eVGA SR-3 2012/01/05 09:51:24
Forgotten can 3960X run in a dual CPU board?? ie two 3960 in same board???
beaker7
Superclocked Member
Re:eVGA SR-3 2012/01/05 09:58:57
pazza3169

Forgotten can 3960X run in a dual CPU board?? ie two 3960 in same board???

 
No.
pazza3169
New Member
Re:eVGA SR-3 2012/01/05 10:11:04
I thought not!! But a shame as I have two 5.1 stable samples here. That clock amazing on water.
lehpron
Regular Guy
Re:eVGA SR-3 2012/01/05 10:42:21
Forgotton
This means ALL of these processors are slotted to be usable by the SrX
This is a very wrong statement.  Core i7-900 has the same 1366 pin-out as Xeons 5500/5600 but doesn't function at all in dual-socket boards.  You can't just make the leap that because i7-3800/3900 have the same 2011 pin-out and overall technical similarity with other 2011's that they are all equally compatible with SR-X.  It is a gross oversimplification that misinforms and you're abusing the fact sheet for it. 
 
pazza3169
Forgotten can 3960X run in a dual CPU board?? ie two 3960 in same board???
This thread originally started out in the X79 forums because somone assumed all LGA2011socket boards were the same and thus the CPU's can be mis-matched-- they can't; and why this thread was moved to the SR-2 subforum.  EVGA knows their own board better than any rumormill
post edited by lehpron - 2012/01/05 11:06:16
pazza3169
New Member
Re:eVGA SR-3 2012/01/05 10:46:02
I am aware of what the board is I was just asking about the rumour created. I have Xeon 8 Core ES in hand anyway. I wish a mod would clear up all the rumours regarding this board and its potential for clocking.
post edited by pazza3169 - 2012/01/05 10:48:46
lehpron
Regular Guy
Re:eVGA SR-3 2012/01/05 11:13:58
EVGA's SR-2 was the only overclockable dual-1366 board, even SuperMicro makes dual-1366 boards with as many PCIe x16 slots, so the only thing EVGA has is the ability to overclock the CPU's.  Therefore, it is very unlikely that SR-X will be castrated, per se, we won't need official say for that.
 
That said, the question is, which Xeons are unlocked; that I don't know.  But I figure similar to the approach with X79, I think only a handful will be able and possibly 8-core only -- I'm betting this since Intel already has unlocked 6-cores in X79 and unlocked quads in LGA1155.  Those that appreciate the appeal of dual-socket the most are those who can afford more, so not someone that purcahsed dual quads in SR-2 for instance.
pazza3169
New Member
Re:eVGA SR-3 2012/01/05 11:25:00
I have the 8 Core top model Xeon on ES and I an tell you its locked. Fully locked.
Forgotton
iCX Member
Re:eVGA SR-X (SR-3) 2012/01/05 14:08:36
 
 
Processor Number Cache Clock Speed # of Cores/
# of Threads Max TDP Memory Types Graphics Intel® Xeon® Processor X5690 (12M Cache, 3.46 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.46 GHz 6 / 12 130 DDR3-800/1066/1333 No Intel® Xeon® Processor X5675 (12M Cache, 3.06 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.06 GHz 6 / 12 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5687 (12M Cache, 3.60 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.60 GHz 4 / 8 130 DDR3-800/1066/1333 No Intel® Xeon® Processor X5672 (12M Cache, 3.20 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.20 GHz 4 / 8 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5647 (12M Cache, 2.93 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.93 GHz 4 / 8 130 DDR3-800/1066 No Intel® Xeon® Processor E5649 (12M Cache, 2.53 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.53 GHz 6 / 12 80 DDR3-800/1066/1333 No Intel® Xeon® Processor E5607 (8M Cache, 2.26 GHz, 4.80 GT/s Intel® QPI) 8.0 MB 2.26 GHz 4 / 4 80 DDR3-800/1066 No Intel® Xeon® Processor E5606 (8M Cache, 2.13 GHz, 4.80 GT/s Intel® QPI) 8.0 MB 2.13 GHz 4 / 4 80 DDR3-800/1066 No Intel® Xeon® Processor E5603 (4M Cache, 1.60 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 1.60 GHz 4 / 4 80 DDR3-800/1066 No Intel® Xeon® Processor X5680 (12M Cache, 3.33 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.33 GHz 6 / 12 130 DDR3-800/1066/1333 No Intel® Xeon® Processor X5667 (12M Cache, 3.06 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.06 GHz 4 / 8 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5670 (12M Cache, 2.93 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 2.93 GHz 6 / 12 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5660 (12M Cache, 2.80 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 2.80 GHz 6 / 12 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5650 (12M Cache, 2.66 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 2.66 GHz 6 / 12 95 DDR3-800/1066/1333 No Intel® Xeon® Processor E5640 (12M Cache, 2.66 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.66 GHz 4 / 8 80 DDR3-800/1066 No Intel® Xeon® Processor E5630 (12M Cache, 2.53 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.53 GHz 4 / 8 80 DDR3-800/1066 No Intel® Xeon® Processor E5620 (12M Cache, 2.40 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.40 GHz 4 / 8 80 DDR3-800/1066 No Intel® Xeon® Processor L5640 (12M Cache, 2.26 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.26 GHz 6 / 12 60 DDR3-800/1066/1333 No Intel® Xeon® Processor L5630 (12M Cache, 2.13 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.13 GHz 4 / 8 40 DDR3-800/1066 No Intel® Xeon® Processor L5609 (12M Cache, 1.86 GHz, 4.80 GT/s Intel® QPI) 12.0 MB 1.86 GHz 4 / 4 40 DDR3-800/1066 No Intel® Xeon® Processor X5677 (12M Cache, 3.46 GHz, 6.40 GT/s Intel® QPI) 12.0 MB 3.46 GHz 4 / 8 130 DDR3-800/1066/1333 No Intel® Xeon® Processor L5638 (12M Cache, 2.00 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.00 GHz 6 / 12 60
No Intel® Xeon® Processor L5618 (12M Cache, 1.87 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 1.87 GHz 4 / 8 40
No Intel® Xeon® Processor E5645 (12M Cache, 2.40 GHz, 5.86 GT/s Intel® QPI) 12.0 MB 2.40 GHz 6 / 12 80 DDR3-800/1066/1333 No Intel® Xeon® Processor W5590 (8M Cache, 3.33 GHz, 6.40 GT/s Intel® QPI) 8.0 MB 3.33 GHz 4 / 8 130 DDR3-800/1066/1333 No Intel® Xeon® Processor L5530 (8M Cache, 2.40 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.40 GHz 4 / 8 60 DDR3-800/1066 No Intel® Xeon® Processor E5502 (4M Cache, 1.86 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 1.86 GHz 2 / 2 80 DDR3-800 No Intel® Xeon® Processor E5506 (4M Cache, 2.13 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 2.13 GHz 4 / 4 80 DDR3-800 No Intel® Xeon® Processor E5530 (8M Cache, 2.40 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.40 GHz 4 / 8 80 DDR3-800/1066 No Intel® Xeon® Processor E5540 (8M Cache, 2.53 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.53 GHz 4 / 8 80 DDR3-800/1066 No Intel® Xeon® Processor X5550 (8M Cache, 2.66 GHz, 6.40 GT/s Intel® QPI) 8.0 MB 2.66 GHz 4 / 8 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5560 (8M Cache, 2.80 GHz, 6.40 GT/s Intel® QPI) 8.0 MB 2.80 GHz 4 / 8 95 DDR3-800/1066/1333 No Intel® Xeon® Processor X5570 (8M Cache, 2.93 GHz, 6.40 GT/s Intel® QPI) 8.0 MB 2.93 GHz 4 / 8 95 DDR3-800/1066/1333 No Intel® Xeon® Processor W5580 (8M Cache, 3.20 GHz, 6.40 GT/s Intel® QPI) 8.0 MB 3.20 GHz 4 / 8 130 DDR3-800/1066/1333 No Intel® Xeon® Processor E5520 (8M Cache, 2.26 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.26 GHz 4 / 8 80 DDR3 1066/800 No Intel® Xeon® Processor L5520 (8M Cache, 2.26 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.26 GHz 4 / 8 60 DDR3-800/1066 No Intel® Xeon® Processor E5504 (4M Cache, 2.00 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 2.00 GHz 4 / 4 80 DDR3 800 No Intel® Xeon® Processor L5506 (4M Cache, 2.13 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 2.13 GHz 4 / 4 60 DDR3-800 No Intel® Xeon® Processor L5508 (8M Cache, 2.00 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.00 GHz 2 / 4 38 DDR3-800/1066 No Intel® Xeon® Processor L5518 (8M Cache, 2.13 GHz, 5.86 GT/s Intel® QPI) 8.0 MB 2.13 GHz 4 / 8 60 DDR3-800/1066 No  
Intel® Xeon® Processor E5503 (4M Cache, 2.00 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 2.00 GHz 2 / 2 80 DDR3-800 No Intel® Xeon® Processor E5507 (4M Cache, 2.26 GHz, 4.80 GT/s Intel® QPI) 4.0 MB 2.26 GHz 4 / 4 80 DDR3-800 No Intel® Xeon® Processor EC5549 (8M Cache, 2.53 GHz, 5.87 GT/s Intel® QPI) 8.0 MB 2.53 GHz 4 / 8 85 DDR3 - 800/1066/1333 No Intel® Xeon® Processor EC5539 (4M Cache, 2.27 GHz, 5.87 GT/s Intel® QPI) 4.0 MB 2.27 GHz 2 / 2 65
No Intel® Xeon® Processor LC5528 (8M Cache, 2.13 GHz, 4.80 GT/s Intel® QPI) 8.0 MB 2.13 GHz 4 / 8 60
No Intel® Xeon® Processor EC3539 (8M Cache, 2.13 GHz) 8.0 MB 2.13 GHz 4 / 4 65
No Intel® Xeon® Processor EC5509 (8M Cache, 2.00 GHz, 4.80 GT/s Intel® QPI) 8.0 MB 2.00 GHz 4 / 4 85
No Intel® Xeon® Processor LC5518 (8M Cache, 1.73 GHz, 4.80 GT/s Intel® QPI) 8.0 MB 1.73 GHz 4 / 8 48
No  
 
Intel® Itanium® Processor 9330 (20M Cache, 1.46 GHz, 4.80 GT/s Intel® QPI) 20.0 MB 1.46 GHz 4 / 8 155
No Intel® Itanium® Processor 9310 (10M Cache, 1.60 GHz, 4.80 GT/s Intel® QPI) 10.0 MB 1.60 GHz 2 / 4 130
No Intel® Itanium® Processor 9320 (16M Cache, 1.33 GHz, 4.80 GT/s Intel® QPI) 16.0 MB 1.33 GHz 4 / 8 155
No Intel® Itanium® Processor 9340 (20M Cache, 1.60 GHz, 4.80 GT/s Intel® QPI) 20.0 MB 1.60 GHz 4 / 8 185
No Intel® Itanium® Processor 9350 (24M Cache, 1.73 GHz, 4.80 GT/s Intel® QPI) 24.0 MB 1.73 GHz 4 / 8 185
No  

All released multi processor cpu's SR-X like SR-2 needs QPI link. my previous post showed fact sheets primarily about single processor server chips. yes still for servers. However if you read carefully through them you can find the links and information that relates to the cpu's for the Multi Processor boards which is what i meant in my third post i did make one mistake only after being frustrated because of responses but still the posts did relate just like this one does. these are all multi cpu ready. look for QPI. if you look at the series numbers. then add to the series you can predict what will come out for the SR-X and general price ranges. the difference between this post and this earlier is i have not gone and gotten a way to get to 2011 socket only information.
post edited by Forgotton - 2012/01/05 14:21:48
beaker7
Superclocked Member
Re:eVGA SR-X (SR-3) 2012/01/05 14:13:13
Forgotton
Processor Number Cache Clock Speed # of Cores/

 
None of these work in SR-X.  If you don't believe us, buy some Itaniums and give it a whirl.  
mosie
Superclocked Member
Re:eVGA SR-3 2012/01/05 15:49:33
An idea for NDA finish date ?
and series release date ?
 
ORB, TIN, ... and auther peaple have certenely a good news for potentiol octo es OC.
crazy1323
New Member
Re:eVGA SR-3 2012/01/05 18:52:52
Last I heard, the E5-2600's are supposed to be released in March. 
 
I would think the board would debut soon after that. 
Forgotton
iCX Member
Re:eVGA SR-3 2012/01/05 19:09:27
They are currently slated to be released basically the same time in march.
prometheus_32
Superclocked Member
Re:eVGA SR-3 2012/01/05 19:14:20
WOOT!!! march
Brocasta
SSC Member
Re:eVGA SR-3 2012/01/05 20:07:30
I'll have to inform my electric utility 6-8 weeks prior to launch so they can upgrade my transformer. Wouldn't want to cause a brownout.
 

kimura
New Member
Re:eVGA SR-3 2012/01/06 00:36:25
Well I'm very glad, though a little upset I didn't know sooner, to find the EVGA boards do support ECC and RDIMMS. That's enough for me to know that I will likely purchase the SR-3 from EVGA. I'm a little upset I didn't know that the SR-2 will run the ECC registered memory because I really did want to purchase the board.

EDIT: 
There was a guy here that I thought was saying that ECC memory runs on all systems, (the system obviously needs to support the technologies to use them) but turns out he was just supporting ECC and Registered memory, just as they deserve, especially these days when not a lot of people find the technologies desirable. Good deal! 
post edited by kimura - 2012/01/06 13:21:43
mosie
Superclocked Member
Re:eVGA SR-3 2012/01/06 17:14:16
I have tested in the asus sabertouth: hynix 8GB 1333 ECC modules but not boot. actualy not compatible.
 
But the 2000 C6 gskill worck fine at 2050 and boot at 23XX.
 
Very need a bios update for xeon.
Jason Eller
New Member
Re:eVGA SR-3 2012/01/16 17:54:49
Any idea as to when this will be out?
lollygag
iCX Member
Re:eVGA SR-3 2012/01/16 18:02:02
My wallets been cocked for a while now... 
Brocasta
SSC Member
Re:eVGA SR-3 2012/01/16 19:46:39
Jason Eller

Any idea as to when this will be out?

 
April, probably. If it's sooner than that, it would be a miracle.
rhussain
FTW Member
Re:eVGA SR-3 2012/01/17 14:30:57
I need to do a build asap. I don't know when this board will be out, but I would like to run a dual cpu configuration with a total of 24 threads. Still looking for the best deal on this.
beaker7
Superclocked Member
Re:eVGA SR-3 2012/01/19 10:28:26
rhussain

I need to do a build asap. I don't know when this board will be out, but I would like to run a dual cpu configuration with a total of 24 threads. Still looking for the best deal on this.

 
If you must build ASAP then SR-2 is the only game in town if you want to overclock.
 
If not, there are many 5520 boards available.
Violence.
FTW Member
Re:eVGA SR-3 2012/01/19 10:58:17
can you overclock the sr-x? processor wise?
pazza3169
New Member
Re:eVGA SR-3 2012/01/19 12:05:42
I thought initially not, or certainly only minimal via BLK. But the odd comment in other forums have led me to believe otherwise. I think you will be able to clock the Xeons via the CPU straps. they certainly won be unlocked via multi but if they boot at the 166 strap who needs multipliers!!!
lollygag
iCX Member
Re:eVGA SR-3 2012/01/19 12:17:56
Yup.. 24 threads @ 4.8.  185bclk  x26.. I hope we we be able to do the same on the SR-X.. I cant wait!
post edited by lollygag - 2012/01/19 12:21:43
pazza3169
New Member
Re:eVGA SR-3 2012/01/19 12:34:12
With SRX it will be 32 threads mate!!! but the BLK manipulation will be much more stringent than SR2. You will have to stay close to the strap settings in numbers for stability. However you will haveupto 34 multi to play with.
lastsamurairick
FTW Member
Re:eVGA SR-3 2012/01/19 16:36:41
Oh Snap 
darkagentj007
New Member
Re:eVGA SR-3 2012/01/22 16:07:38
Is there a release date yet? And full specs?
lehpron
Regular Guy
Re:eVGA SR-3 2012/01/22 17:47:18
darkagentj007
Is there a release date yet? And full specs?
Scroll up to reply #155.  Because SR-X needs Xeon E5-2600 series CPUs, it can't debut before that.  Those CPUs are slated for March 2012, look up those rumors. 
Forgotton
iCX Member
Re:eVGA SR-3 2012/01/22 18:00:29
The board is tentatively slated to be released at the same time as the processors march-12 as lampron say's this is promising news as it came from EVGA employees however anyone who watches the tech industry for any length of time knows there can always be delays. so hope for march but do not be all angry if it's later. Before EVGA started releasing pictures of the next board I and others would have advised people to go for SR-2 but with an official release of the pictures. That has changed that advice at least from me. If you can wait for the next release do, if you need something absolutely 100% before or by March I would advise still going with SR-2 despite it's known issues. If anyone claims its a set date or has given you the impression its 100% sure to come out at the same time of the CPU's or even that the CPU's are sure to come out in march. Take it with a grain of salt and hope for the best but if you can't wait plan for the worst.
nikkocortez
CLASSIFIED Member
Re:eVGA SR-3 2012/01/23 19:28:16
My concern is what the price is going to be.  The SR-2 hasn't budged a bit from $599.99 but when you corner such a market I guess it doesn't need to drop.  I'm just hoping this one will be the same or at least not much more.  The price may be the break between this or going back to an SR-2.  I will have to wait I guess.
Brocasta
SSC Member
Re:eVGA SR-3 2012/01/23 19:36:24
I assume it will be more than the SR-2. Even if it's, say, 800, it's still going to give you more value than the SR-2 which will probably not go down in price. The 1366 Xeons certainly wont go down in price.
badass1982
iCX Member
Re:eVGA SR-3 2012/01/30 13:28:36
Ok guys somebody posted something a while back in the thread (few pages ago think it was pg. 4), that annoyed/confused the hell outta me. 
 
Bear in mind I am getting my info from this site: http://www.cpu-world.com/CPUs/Xeon
 
Now I was under the impression that out of the E5 family (the ones coming out that ARE compatible with SR-X , or at least expected to be compatible with it), that in particular the http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon%20E5-2687W.html E5-2687W was/is the best of the currently planned/announced chips.
 
So i guess my question/frustration is that if this chip is practically IDENTICAL to the http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon%20E5-2690.html E5-2690 barring the fact that the E5-2687W has a HIGHER clock speed, then why do they list the latter as being more expensive and thus better (according to the poster referred to at the beginning of this post), or am i missing something BLATANTLY obvious?.
Gratuitous
FTW Member
Re:eVGA SR-3 2012/01/30 14:10:34
more expensive doesnt always equal better. the wattage is lower on the xx90 so it might run cooler, who knows, good question. I plan on waiting a year for something like this though.
http://techiser.com/10core-intel-ivy-bridgeep-cpu-95w-tdp-24ghz-147116.html
 
Forgotton
iCX Member
Re:eVGA SR-3 2012/01/30 17:41:01
badass1982

Ok guys somebody posted something a while back in the thread (few pages ago think it was pg. 4), that annoyed/confused the hell outta me. 

So i guess my question/frustration is that if this chip is practically IDENTICAL to the http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon%20E5-2690.html E5-2690 barring the fact that the E5-2687W has a HIGHER clock speed, then why do they list the latter as being more expensive and thus better (according to the poster referred to at the beginning of this post), or am i missing something BLATANTLY obvious?.

Also that the one's for the SR-X's will need to have QPI links and will be much more stable built more for computational then for gaming or other such mainstream End User processors. Intel believes most of these chips will be for enterprise level server systems.
lehpron
Regular Guy
Re:eVGA SR-3 2012/01/30 17:47:20
badass1982
So i guess my question/frustration is that if this chip is practically IDENTICAL to the http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon%20E5-2690.html E5-2690 barring the fact that the E5-2687W has a HIGHER clock speed, then why do they list the latter as being more expensive and thus better (according to the poster referred to at the beginning of this post), or am i missing something BLATANTLY obvious?.
It is obvious if you were the intended customer, enthusiasts generally aren't and usually downplay the primary factor: TDP rating.  For the typical Xeon customer, TDP matters just as much as performance since they are usually bought in the hundreds so every last wattage matters.  The slightly slower part is more expensive because it is stuck into a lower TDP rating, more than likely its Vcore is lower, hence premium charge.   For all we know, it maybe one of those Xeons with the unlocked multiplier applicable to SR-X, which would boost the cost anyway.
 
I figure enthusiasts used to the 130W TDP range of X58 and X79 parts would rather maintain that thermal comfort level with an 8-core in the 135W TDP category rather than 150W TDP. 
 
Furthermore, 
  • TDP isn't actual per CPU, it is just a category so Intel can debut future parts within the same category and not exceed that upper limit
  • "T" in TDP stands for "thermal", meaning it is a heat category and not power draw (which is a common misconception because the rate of heat dissipation and the rate of electrical draw happen to be both measured in Watts in the metric system). 
  • Intel will **never** publically publish their power data per CPU, so they hide it by aligning their CPUs in thermal categories to make it easy for cooling companies to design one cooler rated for each TDP that qualifies for any Intel CPU in that category.  As long as there are socket adapters, any 130W TDP cooler works with any CPU rated at 130W TDP or less until the end of time; that's the idea. 
  • Most importantly, if a CPU's actual full load wattage is just above one category, it gets stuck into another; if a CPU dissipated 85W of heat, it will qualify for 95W TDP but not 80W TDP.  Meaning, chances are very high 2687W just barely disqualified for 135W TDP and Intel has enough room to debut faster 8-cores within the same TDP rating.
Short of a calorimeter, the rate of heat dissipation cannot be measured.  Last year I did an guesstimation of sorts, to analyze each TDP and spec of LGA2011 Xeon E5's to figure what the actual TDP wattage could be.  This was my result:  
 
Estimated actual TDP for 32nm LGA2011 parts = 27 + 1.414 x (number of cores) x (core frequency)2
 
This way, the actual wattage of the 2690 becomes 122W of heat dissipation; over the 115W TDP (thus not qualfying) and under 135W TDP (which it does) -- but also 130W TDP, so if you want to tweak the math a bit go ahead, I was trying to get as many Xeons in one formulas as possible.)  Going by this Core i7 3960 ends up around 119W, so the E5-2690 and i7-3960X at stock should have about the same temperatures with the same coolers.  We'll see in a month if that is true.
 
My primary assumption for this formula is that the voltage per core for every Xeon E5 in LGA2011 is the same; more than likely this is wrong so take it with salt.  But this formula makes sense for the majority.  It could even apply to X79 parts, but it does not apply to LGA1155 or LGA1356 parts.  That 27W constant takes into account the 40 lanes of PCIe and quad-channel DDR3 that remains constant for all LGA2011 parts, you'd have to fractionalize for other sockets.  Looking at the SNB-EP die shot, 80% of the surface area was cores and cache; then assumed wattage and surface area were porportional (which they are not) and treated the 2687's actual wattage as barely qualifying for 150W TDP rating setting it to 135W TDP exacty-- then scaled everything other Xeon E5 down.
 
Can you use that equation for overclocking?  Only if you don't touch voltages, because of the primary assumption.  Adjusting voltage has an exponential effect heat and power wattages.  Those of you lucky enough to double the frequency of 8-cores for WR benchmark runs will end up needing over 40A per CPU.  Doubling frequency tends to quadruple the wattage; you were warned.
post edited by lehpron - 2012/01/30 17:55:28
debs3759
New Member
Re:eVGA SR-3 2012/01/30 17:52:11
Any of the E5-2600 processors should work in the SR-X (they are part of the same platform!). The E5-4600 (quad processor capable) chips probably work as  well, but I have yet to see any real news about them other than Intel published part numbers.
Page: < 12345.. > >> Showing page 3 of 7
  • Back to Mobile