• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Ivy Bridge (3770K & 3570K) Results and Discussion thread

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Did you try loading one of the LN2 presets? You have have to set up slow mode if you have cbb
Messing around tonight, had a CBB at -75C and CB at -108C. Tried LN2 mode, various CPU voltages from 1.5V to 1.7V, made no difference. Once it hit -108C at the base of the pot, the system locked up.

Was only running 4.8GHz and modest RAM frequency. Going to try again tomorrow with different CPU PLL settings. Anyone heard lower and upper ranges for CPU PLL? I only went between 1.8V and 1.85V tonight.
 
Messing around tonight, had a CBB at -75C and CB at -108C. Tried LN2 mode, various CPU voltages from 1.5V to 1.7V, made no difference. Once it hit -108C at the base of the pot, the system locked up.

Was only running 4.8GHz and modest RAM frequency. Going to try again tomorrow with different CPU PLL settings. Anyone heard lower and upper ranges for CPU PLL? I only went between 1.8V and 1.85V tonight.

Raising BCLK is the key to avoiding cbb, or so I've heard. From Shammy:

Points to watch when running LN2 Cooling for the best cold-bug free runs:
- Use a BCLK of 105MHz or higher
- Find the right CPU PLL Voltage that your CPU likes (can be higher or lower than default)
- Some CPUs may only initialize their high IMC Frequency at a warmer temperature though they may work fine after that
- ‘23’ when trying to boot means cpu imc not able to run at that frequency at that temperature. ‘30’ means the same and a bit more = ‘23’
- Highest DRAM Ratio known to be working = 2933 ratio
 
I hadn't gotten around to checking on power consumption but I'm currently running AIDA64 stability test 100% load at 45x with my 3570K using the on die graphics along with four fans, two pumps and one hdd and my Kill-a-watt shows pulling just ~135 watts!
 
Last edited:
SR0PL Batch# L206A973 5.72GHz at 1.75V would do wp32 and wp1024:
http://hwbot.org/submission/2279683_
http://hwbot.org/submission/2279684_

Cold boot bugs at -75C, cold bugs around -105C. Tested CPU PLL from 1.7V to 1.95V (default read 1.812 in BIOS). Tested bclk from 105-110, trouble getting higher bclk to post. Tested CPU voltage from 1.0V to 1.75V. None of these items, with or without LN2 mode enabled on the MVG, made any difference at all upon the cold bug behavior of the chip from what I found.

I did not test RAM. Really only tested cold bug behavior, and once I found I couldn't work around it with anything I was trying, I didn't want to press the chip any harder.

Looking to get another from a different store (and hopefully different batch) in the next week or two and will report back.
 
That sucks...and is quite surprising. Certainly hope the next chip does better!
 
SR0PL Batch# L206A973 5.72GHz at 1.75V would do wp32 and wp1024:
http://hwbot.org/submission/2279683_
http://hwbot.org/submission/2279684_

Cold boot bugs at -75C, cold bugs around -105C. Tested CPU PLL from 1.7V to 1.95V (default read 1.812 in BIOS). Tested bclk from 105-110, trouble getting higher bclk to post. Tested CPU voltage from 1.0V to 1.75V. None of these items, with or without LN2 mode enabled on the MVG, made any difference at all upon the cold bug behavior of the chip from what I found.

I did not test RAM. Really only tested cold bug behavior, and once I found I couldn't work around it with anything I was trying, I didn't want to press the chip any harder.

Looking to get another from a different store (and hopefully different batch) in the next week or two and will report back.

Well I have 4 chip and one I know went 5.8 on DICE will be LN2 tomorrow with the Gigabyte board and MSI board :)

Hope one will be 6+
 
SR0PL Batch# L206A973 5.72GHz at 1.75V would do wp32 and wp1024:
http://hwbot.org/submission/2279683_
http://hwbot.org/submission/2279684_

Cold boot bugs at -75C, cold bugs around -105C. Tested CPU PLL from 1.7V to 1.95V (default read 1.812 in BIOS). Tested bclk from 105-110, trouble getting higher bclk to post. Tested CPU voltage from 1.0V to 1.75V. None of these items, with or without LN2 mode enabled on the MVG, made any difference at all upon the cold bug behavior of the chip from what I found.

I did not test RAM. Really only tested cold bug behavior, and once I found I couldn't work around it with anything I was trying, I didn't want to press the chip any harder.

Looking to get another from a different store (and hopefully different batch) in the next week or two and will report back.

I have the same batch, look like MC country wide all got the same batch:rain:
 
I think mine is the same as well, I'll have to check when I get home.

zRVDG.jpg
 
mmm, regarding the batch number: if it's like SBs, it should not be of great importance.
 
I have the same batch, look like MC country wide all got the same batch:rain:

Interesting.

I don't know if batch matters this round or not? I guess I haven't seen enough batches collected anywhere yet, to see if there is some kind of good/bad trend with any of the batches.

Mine is Malaysia as well by the way.

For those keeping score at home, my chip probably could have been worked up to 6GHz with the right voltages and love/care. I didn't work frequencies all that much when nothing I was doing had any effect on the cold bug... Hoping to find another chip that doesn't have the cold bug issues, then will pound on frequencies much harder. Ideally, I'd like to get things colder before I put the screws to it... At these temps I'm competing against anyone with a cascade or even DICE, if I'm pouring LN2 I want to be competing against the other guys with LN2.
 
mine is Costa Rica also SR0PL ... bought in Poland ;)

SS is no fun. -28*C idle on cpu and all I can make is ~5.4GHz for tests :( It's faster clock to clock than SB but I was counting on at least 5.7-5.8GHz :rain:
 
Last edited:
Yeah SS is a waste of time really, you really have to drop the clocks for testing anything that puts a decent load of the CPU. Max I could do on SS was 5.65 for CPUZ and 5.62 for 1m. 3D and wprime more like 5.3-5.4.

Lszoc3.png
 
My 3570k runs at 4400 with 1.26 Vcore
And BSOD x45 even with 1.28 vcore. Thing is according to CPU-Z vcore drops a lot when in Windows and under load. This is on z77-v Pro board and latest BIOS.

Cooling is H80 in well ventilated case (Antec 1200) and temps are not higher than 72C using Prime 27.6

Any ideas on what I could do to improve o/c?
What is a safe 24/7 Vcore for Ivys when using water?
 
Sounds like you need to turn LLC on Melhisedek.

No one can know safe voltages yet without extensive testing for degradation. That said, 2I'd keep Vcore under 1.3V for 24/7 operation, if nothing else than to keep temps in check. If you're feeling bold, anything that keeps it under Tjunction Max (105°C) is probably fair game.
 
Back