Ivy Bridge Discussion, Overclocking/Early Results

April 23, 2012 at 08:56:37
Specs: Windows 7 64Bit, E8200/4GB
I have been reading quite a bit about early results on overclocking the 3570K and the 3770K Ivy Bridge CPU's. There appears to be a few things in common so far:

Clock for Clock the tests appear to be a small improvement over Sandy Bridge.
Most appear to be resorting to higher voltages than expected and this appears to be causing excessive heat within the chip.
This heat build up does not appear to be coming out of the chip itself fast enough so even on Cold water they are having trouble keeping the temperatures under control.
Apparently due to the heat, they are not reaching levels typically that they expected.

Withing a small sampling that I read (confirmed C1 stepping), one 3570K by one person reached 4.6GHz on air at around 1.2Vcore but most 3570k's and 3770K's appear to be requiring 1.4Vcore and higher to get between 4.5GHz and 5.2GHz and even with with water they are getting hotter internally than the owners are comfortable with. One question revealed that in the opinion of the tester, the heat coming out of the radiator was NOT very hot, but was cooler than on tests with other CPU's. This led to a discussion of whether the 'lids' were soldered on or glued onto the chips. However Intel is assembling these, the heat appears to be slow coming out of the chip to the heat sink. These CPU's are all coming out of China and The Netherlands and most were retail boxed.
Is Intel selling weak chips overseas? (does not make sense)
Is Intel going to be revising their chips even now to deal with this (maybe a C2 stepping out later in the summer)? (either better stability at lower voltages or better heat transfer out of the chip)
If these are running hot now, is Intel going to need top and bottom heat sinks with heat pipes coming out of the socket and contacts along the edges on the future 14nm and 10nm chips? Or are they going to conquer running stable at lower voltages at the higher clocks that people want.
Or is the answer in using onchip GPU's only on stock CPU's (for OEM's, businesses, and those who want them) and completely removing them on unlocked CPU's, possibly leaving room for more cores?
Would a diamond coating over the silicon followed by a gold plating, followed by gold soldering a gold plated copper alloy lid help remove the heat?

Thought for discussion....

You have to be a little bit crazy to keep you from going insane.


See More: Ivy Bridge Discussion, Overclocking/Early Results

Report •

#1
April 23, 2012 at 21:27:01
Maybe I got emotional and frustrated (read too many extreme OC posts).
Just found this:
http://www.anandtech.com/show/5763/...
Apparently there are some limits and some heat issues at higher voltages, but we should have expected that the smaller process would run better on lower voltages, but the first ones I read all were assuming that they would need higher voltages for extreme numbers, but they started their bases too high and had trouble even at more moderate OC's. According to the link above, they could have reached the mid 4's with much less voltage and gone from there.
While I am not yet ready to run out and upgrade immediately, I feel much better about the future when I am ready. Again, sorry that I may have gone a bit overboard but discussion is welcome.

You have to be a little bit crazy to keep you from going insane.


Report •

#2
April 24, 2012 at 09:44:01
thats a pretty good topic, regarding the ivy bridge's launch i expected this to happen those cpu's are powerful, i think the forums u read on might have ppl who love to go over the limits in terms of temps and votages...
Intel would not make a mistake to cause their cpu's to stop functioning in less than a year, dont forget that these ivy bridges also support the same socket 1155 but ive heard that ppl with x79 boards wont have the ability to run the cpu at full potential or even might have some issues, while i reccommend ppl to use an x77 board for the ivy bridge cpu's for overclocking and for stability.

computers are a second home
NVIDIA GeForce
GS_Toxict51


Report •

#3
April 25, 2012 at 21:51:03
Look also at this, mainly speculation, but possibly very interesting:
http://www.overclock.net/t/1248454/...

You have to be a little bit crazy to keep you from going insane.


Report •

Related Solutions

#4
April 27, 2012 at 05:13:30
Data sheets
http://www.intel.com/content/www/us...
http://www.intel.com/content/www/us...

You have to be a little bit crazy to keep you from going insane.


Report •

#5
April 27, 2012 at 15:49:14
From Ivy Bridge Datasheet:
Datasheet, Volume 1 69Signal Description 6.9 Error and Thermal ProtectionTable 6-11. Error and Thermal Protection Signal Name Description Direction/Buffer TypeCATERR#Catastrophic Error: This signal indicates that the system has experienced a catastrophic error and cannot continue to operate. The processor will set this for non-recoverable machine check errors or other unrecoverable internal errors. On the processor, CATERR# is used for signaling the following types of errors:• Legacy MCERRs – CATERR# is asserted for 16 BCLKs.• Legacy IERRs – CATERR# remains asserted until warm or cold reset.OCMOSPECIPECI (Platform Environment Control Interface): A serial sideband interface to the processor, it is used primarily for thermal, power, and error management. I/OAsynchronousPROCHOT#Processor Hot: PROCHOT# goes active when the processor temperature monitoring sensor(s) detects that the processor has reached its maximum safe operating temperature. This indicates that the processor Thermal Control Circuit (TCC) has been activated, if enabled. This signal can also be driven to the processor to activate the TCC.Note:Toggling PROCHOT# more than once in 1.5 ms period will result in constant Pn state of the processor.CMOS Input/Open-Drain OutputTHERMTRIP#Thermal Trip: The processor protects itself from catastrophic overheating by use of an internal thermal sensor. This sensor is set well above the normal operating temperature to ensure that there are no false trips. The processor will stop all execution when the junction temperature exceeds approximately 130 °C.

Does this mean they can tolerate 130 °C as a maximum?

You have to be a little bit crazy to keep you from going insane.


Report •

#6
April 28, 2012 at 07:17:37
wow, i really cant believe it..
ive seen a processor that can tolerate up to 80-90 degrees maximum but 130 is another world i mean this could be a theory until someone performs a test to make it a fact.

computers are a second home
NVIDIA GeForce
GS_Toxict51


Report •

#7
April 30, 2012 at 17:57:22
Not worth the upgrade clock for clock if you already have Sandy Bridge, it shows very little improvement on the perfomance desktop side in raw cpu performance. The two decent size improvements come from the gpu and mobile side of things with Ivy Bridge over Sandy Bridge.

My next upgrade will prob be Haswell. it's a new arch which will at least give you 2 years worth of upgrades where Ivy Bridge is a deadend platform and will bring a lot of improvements over the Sandy Bridge arch along with DDR4 support. The GPU side of Haswell should be completely redone from the inside out from what i was reading.

I'm currently using a first gen Core i7 Westmere system and really at this point in time don't see a need to upgrade especially to something that's going to be a deadend platform within a year.

Iron Sharpens Iron.


Report •

Ask Question