Shin Etsu G-751: 5C cooler with no added noise!
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
I bought a bunch of G751 and have it listed for sale for $4.00 each on amdforums.com.
Yes, I'm making a small amount of money. But after paying the PayPal fees, and other costs, and spending hours shipping off lots of Shin Etsu...you get the picture. I'm doing this to help other enthusiasts get this stuff cheap.
http://www.amdforums.com/showthread.php ... did=230316
http://www.amdforums.com/showthread.php ... did=230216
I wil also have some X23 if anyone wants to try that.
Yes, I'm making a small amount of money. But after paying the PayPal fees, and other costs, and spending hours shipping off lots of Shin Etsu...you get the picture. I'm doing this to help other enthusiasts get this stuff cheap.
http://www.amdforums.com/showthread.php ... did=230316
http://www.amdforums.com/showthread.php ... did=230216
I wil also have some X23 if anyone wants to try that.
Last edited by GBR on Sat Jun 21, 2003 10:41 am, edited 1 time in total.
Ex22 wrote:I'd like to try the X23. When and how much will you be selling that for?
btw, a 1 grm syringe, is that one application?
I think that either G751 or X23-7662 could ideally cover 12 to 25 tbred cores. G751 needs to be warmed up and applied carefully to get the desired result.
X23-7662 has an application chemical that allows it to be applied with a viscosity of 100 and thicken over three days to a viscosity of 600....so it is as easy to apply as other compounds.
I sent Ex22 PM.
But, could I ask that emails about purchases be sent to me at VAILCHALET2 at aol.com in the future, so we don't interfere with this thread.
Talking to the engineers, it seems like Shin Etsu X23-7762 may do exactly what you are asking for.LushMD wrote:Halcyon: any follow up report re the new Shin-Etsu compound? I only asked as I would love a compound that combines the efficacy of G-751 with the ease of application of other compounds, such as AS3 (I must admit that it took me forever to properly apply the G-751)? Many thanks. Take care.
-
- *Lifetime Patron*
- Posts: 142
- Joined: Thu May 01, 2003 10:46 pm
- Location: Philadelphia, PA
- Contact:
GBR: I realize that this is the advertised description of the new Shin-Etsu compound, but Halcyon's previous post (which showed a 5 deg. C rise in CPU temp. upon applying the new compound) indicates that this may not be true. Take care.
PS That is to say, it would be wonderful to have a compound that is easier to apply, but not at the cost of 5 deg. C.
PS That is to say, it would be wonderful to have a compound that is easier to apply, but not at the cost of 5 deg. C.
Odd, since the thermal transfer 6W/Mk for X23-7762 vs 4.5 W/Mk on G751.LushMD wrote:GBR: I realize that this is the advertised description of the new Shin-Etsu compound, but Halcyon's previous post (which showed a 5 deg. C rise in CPU temp. upon applying the new compound) indicates that this may not be true. Take care.
PS That is to say, it would be wonderful to have a compound that is easier to apply, but not at the cost of 5 deg. C.
The engineers feel the X23 theoretically should transfer heat better, but the performance is normally quite similar.
I can't offer an opinion based on personal testing yet, I've been too busy ordering and shipping this stuff. I'll know by the weekend.
I've only ordered the G751 and X23-7762, and have just begun using the G751.
I believe that after the forum members who are trying these two compounds run the goop for a while that we will have some insight.
As far as the other compounds , I think that there are more compounds available from Shin Etsu, but the engineer felt that for overclockers we were making a good start with these two compounds.
Their clients are typically big companies like Intel; and they use automated silkscreen appication, or other machine processes. So the other compounds may or may not be the best for our use.
I believe that after the forum members who are trying these two compounds run the goop for a while that we will have some insight.
As far as the other compounds , I think that there are more compounds available from Shin Etsu, but the engineer felt that for overclockers we were making a good start with these two compounds.
Their clients are typically big companies like Intel; and they use automated silkscreen appication, or other machine processes. So the other compounds may or may not be the best for our use.
hitman47 wrote:EDIT: i wanna re-phrase that...
there are 3 versions of Shin-Etsu x-23?...
7798
7783
7762
so:
which of these is the best?
do they all beat g-751?
which are available/will be available/will never be available?
lol thanks all
Update June 17:
G751 seems to be working quite well. Much higher 3Dmark scores and lower Prime95 temps compared to the AS Ceramique I had been running.
Finally hit 17155 3DMark2001 and 5557 3DMark2003!
1700+ @ 2,25Ghz (204 x 11)
Thermalright SLK-900u and Shin-Etsu G-751
Abit NF7-S v2.0
Sapphire Radeon 9500 non-pro 378/621 softmod
2x 256mb Mushkin LevelII PC3200
Maxtor 40GB
Chieftec DW-1 with Austin 420W PSU
LiteOn 16x DVD & 52x24x52 CD-RW
Not bad for $551.00 in parts! .... and lots of help and good advice from forum members!!
G751 seems to be working quite well. Much higher 3Dmark scores and lower Prime95 temps compared to the AS Ceramique I had been running.
Finally hit 17155 3DMark2001 and 5557 3DMark2003!
1700+ @ 2,25Ghz (204 x 11)
Thermalright SLK-900u and Shin-Etsu G-751
Abit NF7-S v2.0
Sapphire Radeon 9500 non-pro 378/621 softmod
2x 256mb Mushkin LevelII PC3200
Maxtor 40GB
Chieftec DW-1 with Austin 420W PSU
LiteOn 16x DVD & 52x24x52 CD-RW
Not bad for $551.00 in parts! .... and lots of help and good advice from forum members!!
Last edited by GBR on Fri Jun 20, 2003 12:19 pm, edited 3 times in total.
My update on Etsu X23-7762.
I couldn't get it to work.
I got constantly 5-6 degrees centigrade higher CPU temps even after "curing" it for a week.
I switched back to Artic Silver III and now got back my lower temps. I'm running XP 2100+ (thoroughbred latest revision) + SLK-800 + 120 mm Papst fan (PLENTY of air) and a 120mm->80mm fan adapter.
I must admit I'm disappointed with this paste. I would have been happy with identical results compared to AS III, but 5 degrees more for a paste that is somewhat expensive, doesn't seem very good to me.
regards,
Halcyon
I couldn't get it to work.
I got constantly 5-6 degrees centigrade higher CPU temps even after "curing" it for a week.
I switched back to Artic Silver III and now got back my lower temps. I'm running XP 2100+ (thoroughbred latest revision) + SLK-800 + 120 mm Papst fan (PLENTY of air) and a 120mm->80mm fan adapter.
I must admit I'm disappointed with this paste. I would have been happy with identical results compared to AS III, but 5 degrees more for a paste that is somewhat expensive, doesn't seem very good to me.
regards,
Halcyon
Halcyon
have you tried an application technique like this?
http://www.amdforums.com/showthread.php ... did=230649
I was using AS Ceramique and found the G751 made a big difference.
I only hit 17182 3D2001 and 5701 3D2003 after changing pastes.
have you tried an application technique like this?
http://www.amdforums.com/showthread.php ... did=230649
I was using AS Ceramique and found the G751 made a big difference.
I only hit 17182 3D2001 and 5701 3D2003 after changing pastes.
Now that orders are arriving, and some people are getting both kinds of thermal compound, here is Shin Etsu's marking system
Red cappped syringe contains X23-7762
Blue cappped syringe contains G751
Another way to tell is by consistency:
-X23-7762 is more like grease and,
-G751 is a bit more like putty.
The thicker G751 is probably the better choice on cores with shims or problems with the contact with the heatsink.
X23 -7762 should be easier to apply and ~might~ offer better performance.
Red cappped syringe contains X23-7762
Blue cappped syringe contains G751
Another way to tell is by consistency:
-X23-7762 is more like grease and,
-G751 is a bit more like putty.
The thicker G751 is probably the better choice on cores with shims or problems with the contact with the heatsink.
X23 -7762 should be easier to apply and ~might~ offer better performance.
GBR,
I have Shin Etsu X-23-7762 (with a label saying: the more viscous version, viscosity 1720).
I warmed it up before applying and it was very thick paste even after warming up.
I didn't use a credit card (as instructed in the page you linked to), but I use a circular motion pressing the heatsink and cpu core together to make a thin and thorough contact.
This method has worked with all the other thermal pastes I've tried before.
If any of you has luck with X23-7762, please post here, because I sure didn't have any luck with it, but I'm willing to try again if everybody else finds it superior to ASIII.
regards,
Halcyon
I have Shin Etsu X-23-7762 (with a label saying: the more viscous version, viscosity 1720).
I warmed it up before applying and it was very thick paste even after warming up.
I didn't use a credit card (as instructed in the page you linked to), but I use a circular motion pressing the heatsink and cpu core together to make a thin and thorough contact.
This method has worked with all the other thermal pastes I've tried before.
If any of you has luck with X23-7762, please post here, because I sure didn't have any luck with it, but I'm willing to try again if everybody else finds it superior to ASIII.
regards,
Halcyon
I'll have to ask the engineer at Shin Etsu, perhaps your compound was made for a specific company's order.halcyon wrote:GBR,
I have Shin Etsu X-23-7762 (with a label saying: the more viscous version, viscosity 1720).
I warmed it up before applying and it was very thick paste even after warming up.
I didn't use a credit card (as instructed in the page you linked to), but I use a circular motion pressing the heatsink and cpu core together to make a thin and thorough contact.
This method has worked with all the other thermal pastes I've tried before.
If any of you has luck with X23-7762, please post here, because I sure didn't have any luck with it, but I'm willing to try again if everybody else finds it superior to ASIII.
regards,
Halcyon
The X23-7762 I ordered has a viscosity of 100 and thickens to 600 over 3 days of use, and thermal transfer goes from 4.23 W/M k to 6.5 W /M k
Interesting numbers - Arctic Silver 3 is rated at 9.0 W/mK - which actually means it is a better conductor of heat than pure mercury - but surely that suggests AS3 is going to outperform the Shin Etsu compound even after curing? Certainly I haven't been very impressed with the G751 stuff - temps are stiill higher than with AS3 after 2 days - another day to go before the game ends, however.GBR wrote:
The X23-7762 I ordered has a viscosity of 100 and thickens to 600 over 3 days of use, and thermal transfer goes from 4.23 W/M k to 6.5 W /M k
I've noticed the AS# thermal transfer numbers as well...but since it is a bench test to generate the thermal transfer data perhaps it doesn't tell the whole story. And silver or copper should transfer heat better.davsmith wrote:Interesting numbers - Arctic Silver 3 is rated at 9.0 W/mK - which actually means it is a better conductor of heat than pure mercury - but surely that suggests AS3 is going to outperform the Shin Etsu compound even after curing? Certainly I haven't been very impressed with the G751 stuff - temps are stiill higher than with AS3 after 2 days - another day to go before the game ends, however.GBR wrote:
The X23-7762 I ordered has a viscosity of 100 and thickens to 600 over 3 days of use, and thermal transfer goes from 4.23 W/M k to 6.5 W /M k
I'm curious, are the core and heatsink lapped? And what are the specifics of your system?
Were you able to apply the G751 in a thin coat?
I'm trying to get a handle on when G751 is the best choice. I think that the size of the molecules in a thermal compound and the phsical qualities and tolerances of the surfaces it is applied to, and the thickness of the thermal compound layer........... may be some of the most important variables.
I think I did a pretty good job on the coating - warmed up the compound and then used a credit card to spread it. Given its consistency I couldn't imagine getting it much thinner.GBR wrote:
I'm curious, are the core and heatsink lapped? And what are the specifics of your system?
Were you able to apply the G751 in a thin coat?
I'm trying to get a handle on when G751 is the best choice. I think that the size of the molecules in a thermal compound and the phsical qualities and tolerances of the surfaces it is applied to, and the thickness of the thermal compound layer........... may be some of the most important variables.
Heatsink is a SLK-900u and the CPU is a 3.06 GHz P4 CPU - fitted in a Sonata case. Neither CPU or heatsink have been lapped - too much effort and risk to the CPU in my opinion. I wouldn't have bothered trying the G751 compound if it only worked with lapped surfaces.
I think its the other way around....G751 is good at filling in and dealing with 'rough' surfaces.davsmith wrote:I think I did a pretty good job on the coating - warmed up the compound and then used a credit card to spread it. Given its consistency I couldn't imagine getting it much thinner.GBR wrote:
I'm curious, are the core and heatsink lapped? And what are the specifics of your system?
Were you able to apply the G751 in a thin coat?
I'm trying to get a handle on when G751 is the best choice. I think that the size of the molecules in a thermal compound and the phsical qualities and tolerances of the surfaces it is applied to, and the thickness of the thermal compound layer........... may be some of the most important variables.
Heatsink is a SLK-900u and the CPU is a 3.06 GHz P4 CPU - fitted in a Sonata case. Neither CPU or heatsink have been lapped - too much effort and risk to the CPU in my opinion. I wouldn't have bothered trying the G751 compound if it only worked with lapped surfaces.
I'm giving my Shin Etsu X-23-7762 a second try. Just applied it today carefully with my ID card as per instructed in the link given above.
I'll let it "cure" for a week and compare results with a 2.2GHz Athlon XP (SLK800 + Papst 120mm) against Artic Silver III.
Expect more results in a week.
One thing in favour of AS III IMHO is that you can appreciate the results immediately. No 'curing' needed
regards,
Halcyon
I'll let it "cure" for a week and compare results with a 2.2GHz Athlon XP (SLK800 + Papst 120mm) against Artic Silver III.
Expect more results in a week.
One thing in favour of AS III IMHO is that you can appreciate the results immediately. No 'curing' needed
regards,
Halcyon
Ok, I have now let the second application of Shin Etsu X-23-7762 (stiffer version, see above) 'cure' for one week.
I got a degree (or perhaps two) drop in temp compared to non-cured temps.
However, the temps are still 6 degrees higher than with Arctic Silver III. I have now gone back to ASIII and am down to my normal temps.
I must say that for me, on my system (currently Athlon XP at 2.2 GHz, SLK-800, Papst 120mm fan and a 120->80mm adapter) the Shin Etsu in question is clearly inferior to Arctic Silver III.
YMMV, but I was disappointed, although it was worth testing.
regards,
Halcyon
I got a degree (or perhaps two) drop in temp compared to non-cured temps.
However, the temps are still 6 degrees higher than with Arctic Silver III. I have now gone back to ASIII and am down to my normal temps.
I must say that for me, on my system (currently Athlon XP at 2.2 GHz, SLK-800, Papst 120mm fan and a 120->80mm adapter) the Shin Etsu in question is clearly inferior to Arctic Silver III.
YMMV, but I was disappointed, although it was worth testing.
regards,
Halcyon
halcyon wrote:Ok, I have now let the second application of Shin Etsu X-23-7762 (stiffer version, see above) 'cure' for one week.
I got a degree (or perhaps two) drop in temp compared to non-cured temps.
However, the temps are still 6 degrees higher than with Arctic Silver III. I have now gone back to ASIII and am down to my normal temps.
I must say that for me, on my system (currently Athlon XP at 2.2 GHz, SLK-800, Papst 120mm fan and a 120->80mm adapter) the Shin Etsu in question is clearly inferior to Arctic Silver III.
YMMV, but I was disappointed, although it was worth testing.
regards,
Halcyon
This might illustrate how compounds can perform differently on different systems...and there is no clear theory to identify the variables that cause this.
The one thing that concerns me is that your version of X23-7762 is not the same as the one we are trying. So ~perhaps~ that may have something to do with this.
If you can stand testing some more , I may send you some X23-7762 or G751 to compare. Let me know.
you can eme at vailchalet2 at aol.com