- Joined
- Sep 19, 2015
It is the first time since Ivy Bridge years that I see first time builders asking on how to de-lid on this very forum. Never seen it before.
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
photos of delidding the i9 7920X
- Thread starter batboy
- Start date
It is the first time since Ivy Bridge years that I see first time builders asking on how to de-lid on this very forum. Never seen it before.
It's not because they need it ... they read BS threads and they think it's required, one friend asked me if I can give him delidding tool for SB and he is not even overclocking much above stock ... all because he saw that somewhere
My point is only that those who are delidding are making it on unlocked i5/i7 CPUs. These series are really low % of all CPU sales. Even lower % are in hands of those who are delidding. Based on that ( and stores sales ) I can say that overall sales of delidding tools are really low so manufacturers have to bump prices so it will give them any reasonable profit.
My point is only that those who are delidding are making it on unlocked i5/i7 CPUs. These series are really low % of all CPU sales. Even lower % are in hands of those who are delidding. Based on that ( and stores sales ) I can say that overall sales of delidding tools are really low so manufacturers have to bump prices so it will give them any reasonable profit.
- Joined
- Sep 19, 2015
I totally agrre with you on the first part of your post.
Regarding the second part, if orders have been increased by a x3 factor, eveb if it is still marginal, price increases. Not an absolute, but a relative tag
Ye old school computing thread......
http://www.overclockers.com/forums/showthread.php/785377-The-Good-old-days-of-computing
http://www.overclockers.com/forums/showthread.php/785377-The-Good-old-days-of-computing
- Joined
- Mar 7, 2008
The tool should arrive by lunchtime according to tracking. I put the order in last Tuesday, had dispatch notification on Wednesday, with delivery ending up scheduled for Saturday. Since I picked my work address I had to request they delay until today so someone will be there. Although I want to do a proper job on it, I think I can't resist doing a delid tonight to see what difference it makes.
- Joined
- Mar 7, 2008
Delid and LM done. On 7800X, at my 4.3 GHz 1.1v OC, running 12x 64k FFT on Prime95 (over an hour to allow water loop to warm up), I got:
Before, after, drop:
Max
73 63 10
73 68 5
76 65 10
77 72 5
74 68 6
75 66 9
Core spread (hottest - coolest) before 4, after 9
Average
68 60 8
70 66 4
73 63 10
73 68 5
71 66 5
73 63 10
Core spread before 5, after 8
Ambient before 21.8 after 22.2 so not significant change.
A mixed bag there, I wonder if there is either something uneven about my LM application, or non-flat seating of IHS on die. Temps went down 5C on both hottest core max and average but a bit variable overall. I'm thinking of moving this back into its old case later on to improve water cooling capacity, and can look at the application again at that time.
Side note: I looked at VRM temps afterwards only, it logged a peak of 90C as reported by mobo. When should I get concerned?
Before, after, drop:
Max
73 63 10
73 68 5
76 65 10
77 72 5
74 68 6
75 66 9
Core spread (hottest - coolest) before 4, after 9
Average
68 60 8
70 66 4
73 63 10
73 68 5
71 66 5
73 63 10
Core spread before 5, after 8
Ambient before 21.8 after 22.2 so not significant change.
A mixed bag there, I wonder if there is either something uneven about my LM application, or non-flat seating of IHS on die. Temps went down 5C on both hottest core max and average but a bit variable overall. I'm thinking of moving this back into its old case later on to improve water cooling capacity, and can look at the application again at that time.
Side note: I looked at VRM temps afterwards only, it logged a peak of 90C as reported by mobo. When should I get concerned?
OP
- Joined
- Jan 12, 2001
- Location
- Kansas, USA
- Thread Starter
- #30
The results from your delidding and use of LM on the core/inner IHS seems consistent with what I saw when I did mine. I'm thinking maybe these later Skylake-X processors are getting better TIM application from Intel than previously?
- Joined
- Aug 20, 2003
Hah, I keep thinking about it with my 7920x. I do have a slight problem, USPS seems to have lost my 7920x in transit... Was supposed to be here Monday but tracking hasn't been updated since...
- Joined
- Aug 20, 2003
Well USPS came through yesterday:
- Joined
- Aug 20, 2003
Better late than never!
True, I figured I may not see that CPU ever. I've only had that happen a few times with USPS... Now I just need to find time to see how it overclocks.
- Joined
- Aug 20, 2003
What are you running it at?
I'm using it mainly for tests/reviews but recently I was switching platforms and I was even playing games ... It runs at 4.5GHz 1.15V for most of the time. I'm using 2x360 slim rads but it's barely ever under 100% load.
btw. 4000 17-17-17 / Patriot Viper4 3733 on Taichi XE and at 3400MHz cache. bandwidth is a bit lower than I was expecting but it's maybe because of early BIOS
... It runs at 4.5GHz 1.15V for most of the time. I'm using 2x360 slim rads but it's barely ever under 100% load.btw. 4000 17-17-17 / Patriot Viper4 3733 on Taichi XE and at 3400MHz cache. bandwidth is a bit lower than I was expecting but it's maybe because of early BIOS
- Joined
- Aug 20, 2003
I'm using it mainly for tests/reviews but recently I was switching platforms and I was even playing games
btw. 4000 17-17-17 / Patriot Viper4 3733 on Taichi XE and at 3400MHz cache. bandwidth is a bit lower than I was expecting but it's maybe because of early BIOS
View attachment 194781
Dang nicely done. Your L3 is much much faster than mine. Are you adding volts to the mesh to get it to overclock so much?
1.2V mesh but something isn't stable at these settings. It passed about 20 mins AIDA64 so it's not bad. On the other hand on ASUS mobo I couldn't even run tests above 3200.
L3 is related to couple of things like cache frequency but also some others. When you have too many power saving options enabled then cache bandwidth will be lower.
L3 is related to couple of things like cache frequency but also some others. When you have too many power saving options enabled then cache bandwidth will be lower.
