Shin-Etsu G751, X23-7762, and X23-7783D Review


This is the ninth installment of many of the Skinnee Labs TIM Comparison 2011.  For most of the installments, results will be released in two or three TIM sets roughly every two or three weeks and today we have three TIMs to show.

Today we have three pastes from one of the biggest names in TIM: Shin-Etsu.  Shin-Etsu is mostly a supplier to the Rich and Famous of the PC world; we know Apple is a big customer of theirs and suspect the likes of HP and Dell are as well.  In today’s review, we’re testing their most popular (and acquirable) pastes in the retail market, G751, X23-7762, and X23-7783D (which has already been tested on our old testbed and impressed). There are three newer varieties we’re also interested in: G765, X23-7868-2D, and X23-7921-5, but those will have to wait as none seem to be available in retail.  Shin-Etsu pastes can also be found as rebrands while maintaining the same model numbers.  In this case, G751 was found under the Masscool brand and X23-7762 was found under the Cooljag brand.  X23-7783D was found under the actual Shin-Etsu brand and G751/X23-7762 can be as well, with enough searching.

After today, we’ll have 26 TIMs completed in the 2011 Comparison, but we still have over 20 more on the docket.  Lots of testing left to do!

The TIMs

Shin-Etsu G751 is the lowest spec’d paste of the three and was found under the Masscool brand. It’s a medium-high thickness gray paste that is slightly more wet than its two siblings.  I was only able to find it in one size: .5g for $5. Due to the small quantity supplied, I was only able to average five mounts for every four syringes, carrying a very unattractive $4 per mount cost.  So while $5 isn’t a bad entry price, $10 a gram is an awful value without considering performance.

Shin-Etsu X23-7762 is the middle brother of the lineup, according to the specs. It’s a dry, medium-high thickness, gray paste that spreads decently under pressure. X23-7762 seems to be the easiest to find in ‘other’ brands and the paste for this review was of Cooljag branding. At $7 for 1.5g, it has neither a great entry price ($7) nor great value per gram (~$4.67/g). To make matters worse, I only averaged two mounts per syringe, making the price per mount a very unattractive $3.50. X23-7762 can be found at other sizes and price points from other brands with enough searching, but none seem like any better of a value.

Shin-Etsu X23-7783D was, until recently, Shin-Etsu’s best paste and has the best specs of the three pastes in this test. It’s a medium-high thickness gray paste that is between X23-7762 and G751 for moistness. X23-7783D does spread pretty well, which is important.  In comes in two reasonable sizes (and an unreasonable 1kg): .5g and 1g for $4 and $7, respectively.  $4 is a decent entry price but $3.50/g is a poor ‘best’ value.  However, I was able to get three mounts from each 1g syringe, so it is a better price per mount than G751 or X23-7762 in addition to being the better spec’d paste. Without considering performance, X23-7783D’s value is very poor relative to other pastes on the market, however.

The Blocks

For all three contact “settings” I use a Koolance CPU-360.  I’ve chosen the CPU-360 due to its great mounting system (although I’ve modified all three blocks’ mounting systems) and because it’s pretty easy to add slight modifications.

At the “Poor” end of the spectrum, I have a stock CPU-360r1.2 with extremely low mounting pressure; the stock CPU360r1.2 has a somewhat irregular base and when paired with low pressure, TIM does not spread into a thin layer particularly well.

For the “Moderate” contact setting, I’ve taken a CPU-360r1.1 and reduced some of the internal structure so that there’s absolutely no bow.  With pressure in the center of the base the block can actually become slightly concave as the o-ring compresses, but with only moderate mounting pressure the base seems to stay perfectly flat.

As for “Great” contact, I might have gone a little overboard; no block on the market has contact this great.  I’ve modified a CPU-360r1.2 to have a thicker midplate with a compressible layer and the result is a pretty extreme bow that flattens with mounting pressure.  And there’s a lot of mounting pressure.  The result is impressive; with low viscosity pastes, it looks like there isn’t even any TIM on the center of the CPU when taking the mount apart.  Even with viscous pastes the resulting layer is extremely thin.

Pages: 1 2 3 4 5


Posted On
Aug 27, 2011
Posted By

one thing i do not under stand from skineelabs benchmark is
when i performed my own bench mark , better contact led to
less temperature difference no matter what thermal compound i used.

but skineelabs benchmark is totally refreshing on every contact
surface every time.

i do not trust your results .

Posted On
Aug 28, 2011
Posted By
Eric (Vapor) Hassett

Great contact has the lowest difference between best/worst TIMs and Poor contact has the biggest difference between best/worst TIMs and Moderate contact is between the two.

Looking at it from a different angle, in 76 of 78 potential improvement cases (26 TIMs * 3 cases: Moderate vs. Poor, Great vs. Moderate, and Great vs. Poor), the TIMs improved.

With OCZ Freeze and Deep Cool Z9, I’m not sure why they got worse with Great contact compared to Moderate contact, but they did. The reality is that there’s no guarantee that Great (or Moderate) contact will be better than the ‘worse’ contact setting(s)–Great/Moderate/Poor are just names I have assigned to the three settings.

Great contact is a different type of contact than Moderate contact (bowed + very high mounting pressure vs. flat + moderate-high mounting pressure, respectively). It looks like the best contact with OCZ Freeze and Deep Cool Z9 is not using a bowed block with very high mounting pressure. Maybe it’s a bowed block with moderate-high mounting pressure, a flat block with very high mounting pressure, our Moderate contact setting, or something else entirely. No matter what the best is, it does appear that OCZ Freeze and Deep Cool Z9 have degraded performance from too much mounting pressure with the Great contact setting.

As for how we test, because Great/Moderate/Poor contact are all different types of contact (not just different mounting pressures), we use three blocks (each with their own mounting pressure) and do five mounts per setting. This means 15 mounts with no ‘reused’ mounts. Even if we had two contact settings using one block, we couldn’t ‘reuse’ a mount at a different mounting pressure because that would prevent any useful cure measurement.

Looking at Shin-Etsu’s X23-7921-5 data sheet, it appears that with X23-7921-5 higher mounting pressure does not necessarily correlate to improved performance (or even reduced bondline thickness). There is such a thing as too much mounting pressure, it looks like OCZ Freeze, Deep Cool Z9, and even X23-7921-5 readily exhibit degraded performance from significant increases in mounting pressure.

Posted On
Aug 19, 2011
Posted By

Always good information to be found in these TIM tests. Thanks for the massive amount of work you’re putting into this. I totally appreciate it

Posted On
Aug 18, 2011
Posted By

Hi there Skinnee Labs Team!

Thank You for all the good work you do in here mates :-)

As a token of my thankfulness I make some advertising for Skinnee Labs on my FaceBook and “WOLF High Performance Computer Components & Accessories” as showing my computer components as well as upgrades.

WOLF High Performance Computer Components & Accessories. (look further down just after ProlimaTech PK-1 TIM’s).

By this I wish you all at Skinnee Labs best of future to come.

Regards, Bengt “WOLF” Johansson.

Posted On
Aug 18, 2011
Posted By

Still no IC Diamond?

Posted On
Aug 18, 2011
Posted By
Eric (Vapor) Hassett

Got a looooong way to go before IC Diamond gets tested. Can’t test it mid-stream because of how much it could alter the surfaces and therefore alter the performance (before testing officially started I did a trial run and it took just two mounts of IC Diamond to remove Intel’s markings from my IHS…15 mounts could seriously change things).

Posted On
Aug 18, 2011
Posted By

What’s there to say, if you DON’T use Indigo Xtreme… you might as well use Sour Cream !!! Hey that rhymes. But all kidding aside, I STILL don’t understand WHY people still would use TIM pastes anymore. Yeah for short term testing that’s cool, but for long term use I don’t know why you would want to change TIMs every 4 to 6 months. Plus the lower temp scores extend the life of your core processor.

I mean if you’re not wanting to use the Indigo Xtreme (IX) ETI, because of the time it takes to install it, then at least use CoolLab’s Liquid Ultra. At least that’s still a metal to metal contact TIM, and it’s kind of a go-between from using both the IX & a regular TIM. Plus, I don’t think you have to change that metal TIM but once a year. Fantastic alternative and gives almost the same lower temps as the IX.

Even with the issue of the lengthy install with the IX, you won’t have to change that bad boy for at least 2 years. I’ve had mine on for a FULL YEAR on my i7 930 @ 4.2GHz and it STILL stayed at 39C in a 81F room, and that’s water-cooled. Now I have a i7 970 @ 4.2GHz @ 30C in a 81F room. Not to shabby. Just an amazing ETI…

Thanks again for your amazing and thorough reviews !!!

Posted On
Aug 18, 2011
Posted By
Eric (Vapor) Hassett

There’s a lot of reasons to not use IX, chief among which are compatibility and value.

PK-1 is less than a dollar a mount, less than degree behind IX at most heatloads with at least moderate contact, and works on all CPUs, GPUs, RAM, VRMs, etc.

IceFusion is just pennies (maybe not even plural, very little is needed because it spreads so well) a mount and works on all CPUs, GPUs, RAM, VRMs, etc.

Then there’s MX-2 between those in both cost and performance with full compatibility with all CPUs, GPUs, RAM, VRMs, etc.

Point is, depending on how much an individual values a dollar and a degree, IX isn’t always the most attractive. If you want to squeeze out every last degree, IX is awesome.

Leave a Reply

You must be logged in to post a comment.