ICD 7 Test Results

[Fardringle]

Originally posted by: ICD7
"""Using that:
"turn screw until paste appears, then 2 more full turns, in a blob and stop. Then press and twist heatsink using small rotations, many times. Then secure HSF using manufacturers retention device. Boot and watch temps". """

Referring to Fardringle's post the application procedure needs a little "something" after the squish part. Alot of users complain if they try to spread the paste with a razor blade or finger in a baggie, you think it's enough to put in extra twists?

I didn't bother trying to be precise with it like I would with Arctic Silver, I just used an old credit card to spread the paste around a bit (sort of like spreading butter on toast), and then used the "squish and twist" method to finish it off. I didn't bother taking the heat sink off again after that to see if it covered 100% of the core since the mounting bracket on my Thermalright SI-128 is a bit of a pain, but the temperature readings improved dramatically when I did that instead of trying to squish the paste out from a single blob in the center of the CPU.

 

[Insidious]

Originally posted by: ICD7
On application size

We came up with the pea thing when we analyzed some of the first results we got on our first giveaway and on the subsequent HardwareLogic giveaway overall user results improved almost .75 C with the use of a pea sized call out.

Smaller than a breadbasket but larger than a grain of rice...We realize it probably is a little overkill but want to insure that people get a good result, We are trying to remove the subjective guesswork out of application.

One Idea we had is to put a window on a graduated syringe and call out something like 1/10 of a CC per application.

Another is we have this nifty screw type syringe which we can mark a specified amount by 1/4 turn 1/2 turn whatever it works out to be.

Anybody else have a method that might work? let me know we are open to suggestions.

Also to note
Diamond weight vs silver weight-It is easy to forget weights and volumes. AS5 is 3.5 grams silver and fills 1/2 cc on a syringe. ICD7 is 1.5 grams and fills 1/2 cc on syringe - - a carat is .2 gram. For OEM and system builders we are going to be selling a 24 carat tube in the near future.

I really like the idea of a dispenser that meters the amount of stuff. I think a screw tube is a great idea! Being confident that I put on the right amount of stuff wouuld be a plus and you would make big points if your dispenser did that for me! (great advertising point: "we take out the guesswork")

I didn't think about the weight/volume differences in different products. A lot of mystery in our selection process (jeeze, there are a lot of choices) could be simplified if this was marketed by number of applications instead of (or along with) grams.

Does ICD7 have a long shelflife? If I bought a big tube, will it be OK laying around for a long time (months) before it's next use?

Referring to Fardringle's post the application procedure needs a little "something" after the squish part. Alot of users complain if they try to spread the paste with a razor blade or finger in a baggie, you think it's enough to put in extra twists?

Maybe include a little gizmo (like those coffee stirs that are shaped like a tiny canoe paddle) and indicate in the instructions that they can use this tool to spread the compound if they desire but point out it isn't normally necessary because of your metering dispenser and the special viscosity of your product.

-Sid

 

[ICD7]

"""I really like the idea of a dispenser that meters the amount of stuff. I think a screw tube is a great idea! Being confident that I put on the right amount of stuff wouuld be a plus and you would make big points if your dispenser did that for me! (great advertising point: "we take out the guesswork")

I didn't think about the weight/volume differences in different products. A lot of mystery in our selection process (jeeze, there are a lot of choices) could be simplified if this was marketed by number of applications instead of (or along with) grams.

Does ICD7 have a long shelflife? If I bought a big tube, will it be OK laying around for a long time (months) before it's next use?

Referring to Fardringle's post the application procedure needs a little "something" after the squish part. Alot of users complain if they try to spread the paste with a razor blade or finger in a baggie, you think it's enough to put in extra twists?

Maybe include a little gizmo (like those coffee stirs that are shaped like a tiny canoe paddle) and indicate in the instructions that they can use this tool to spread the compound if they desire but point out it isn't normally necessary because of your metering dispenser and the special viscosity of your product."""

_________________________________________________________________-


We looked at alot of dispensing Ideas - Stamping, rolling, a cellophane kind of pad thing where you peel off one side stick it to the cpu then peel off the other, silkscreen. etc. We looked to the cosmetics industry as they have some pretty nifty dispensing systems, but could not find anything applicable. still kind of feeling our way around with it and are open to any suggestions.

I have had stuff sitting around the office for about 20 months with no separation,dryout basically the same as when it went into the tube. I quote shelf life of 2 years to the OEM's because they need a number. when something changes in the first packaged sample then I will post a final shelf life number. I check them once a month but I would not be surprised if it was 3 to 5 years the stuff is pretty stable.

 

[ICD7]

""""Using that:
"turn screw until paste appears, then 2 more full turns, in a blob and stop. Then press and twist heatsink using small rotations, many times. Then secure HSF using manufacturers retention device. Boot and watch temps".""""



twist heatsink using small rotations, many times Looks like we have to add that other step for a somoother application. Anybody else have an opinion?

 

[alfa147x]

Motherboard : GIGABYTE GA-P35-DS3L LGA 775 Intel P35
CPU: Intel Core 2 Duo E8400 Wolfdale
OC = 9 * 417 MHz
HSF = Stock
Ambient = 33ºc
Temps taken using coretemp 0.97

Cheap Thermal Grease
Stock temps:
@3.0GHz
LOAD 64/60? C

Overclock temps:
@3.0GHz
LOAD 65/61 C

ICD7
Stock temps:
@3.0GHz
LOAD 62/57? C

Overclock temps:
@3.0GHz
LOAD 64/59 C


These were taken running 2 F@H clients, sorry i forgot to take idle temps
Thanks a bunch, this stuff kicks ass !!!

 

[narzy]

I ran out of sample on my second attempt with the stuff so I washed it off and installed AS 5, I don't want to submit my results because I think they would skew it, although when I pulled the CPU and heatsync apart it looked like pretty good coverage evenly spread so shrugs, didn't work in my case but I'm not counting the stuff out 🙂 I just found it very thick and hard to work with, if you try to spread it out it becomes thicker and likes to stick together.

 

[Fardringle]

I think they want everyone's results even if their results are "bad", Narzy. In my case, my initial results were very poor but when I re-applied the paste more evenly on my CPU the temperatures improved dramatically.

Even if you weren't able to get good results at all, the point of a test like this is to find out what the results are and to find and (possibly) eliminate the causes of any bad results like yours.

 

[ICD7]

Exactly, This is a performance based beta test, we are looking for trends in those did not receive optimal results as well as those that did. This helps helps us refine our product message, application, troubleshooting etc. Good bad or indifferent, what it is, is what it is.

 

[narzy]

good points, ok I'll update my original post w/ the data.

 

[Uppsala9496]

Just received my sample today so I haven't had a chance to test it yet. Figured I would post what I currently have an update it in the future:
_______________________________________________________
System 1 WinXP Pro:
Motherboard: Asus M2A-VM
CPU: A64 X2 4200+ (stock 2.2ghz)
HSF: Stock
Ambeint: No idea - run F@H nonstop

Stock thermal grease that shipped with retail CPU:
Load: 42-43*C
temp via PC Probe II V.1.04.12
Load: 40*C -24+ hours after application.

Per speedfan (not familiar with so I will post everything):
................................... ICD7 2 hours after application
Temp1: 42C.............. 38-39C......................+ hours24........41C
Temp2: 28C.............. 27C..................................................29C
Temp3: 25C.............. 25C..................................................25C
Temp1: 40C.............. 40C..................................................40C
Core: 42C................. 39-40C..................................................40C ........... +72 hours 38C
_______________________________________________________
System 2 (just got up and running tonight) Linux - Ubuntu:
Motherboard: ECS AMD690GM-M2
CPU: A64 X2 4000+ (oc'd to 2.26ghz)
HSF: Stock
Ambient: unknown - just installed F@H

Stock thermal grease that shipped with retail CPU:
Load: 40*C

temp via sensors-applet 1.7.12

_______________________________________________________
System 3 WinXP Pro:
Motherboard: Foxconn MCP61M2MA
CPU: A64 X2 4000+ (stock 2.31ghz)
HSF: Stock
Ambeint: unknown

Stock thermal grease that shipped with retail CPU:
Load: see below

temp via speedfan
................................... ICD7 2 hours after application
Temp1: 63C.............. 59-60C......................+ hours24........60-61C ........... +72 hours 57-58C
Temp2: 34C.............. 32C..................................................33C
Temp3: 25C.............. 25C..................................................25C
Temp1: 40C.............. 40C..................................................40C
Core: 23C.............. 31C..................................................32C
_______________________________________________________

The room temperature has remained a constant 73-77*F throughout all tests.

 

[biodoc]

Computer #1

HP Pavilion m8120
Asus Basswood3G MB
Q6600 B3 stepping @2.4 GHz
Masscool 8WA741 HSF
Arctic Silver 5 applied 1 month ago
Ambient temp = 68F
Temps monitored with coretemp ver 0.95.4
OS: Vista Home
POEM@HOME Load test

All core temps in Celsius.

Before:

idle: 38 38 38 36
load: 56 56 54 54

After: (30 min post boot for idle) (2 hours at full load)

idle: 35 36 35 33
load: 56 56 54 54

24 h after application:

load: 55 54 52 52 🙂

Application: center pea sized; No "twist & shout" with this HSF because of design and very tight space constraints.

Seems just as good as AS5 with this stock rig.

Edit: Seems a degree or 2 better after 24h!:beer:

I'll take another reading tomorrow morning when ambient is 68 F again.

Reminder to testers: Ambient (room) temp should be the same in before/after tests.

I have another to test (Q6600@3.2 GHz) later today.

Edit: add 24 h post application #s

Computer #2

CPU: Q6600 G0 @3.24 GHz (9x360)
MB: Gigabyte GA-P35_DS3L
vCore: 1.3 volts
RAM: 800MHz
HSF: Arctic Cooler Pro 7
Arctic Silver 5 applied for 2 days
Temps monitored with lm-sensors
OS: ubuntu 64 bit linux
Load Test: 2 x FAH SMP
Ambient temp: 54F (it's cool in my basement)

All core temps in Celsius.

before:

idle: 25 24 24 23
load: 45 42 40 42

after: (2 hours post application for full load)

idle: 22 21 21 20
load: 40 37 37 37🙂

I like this stuff!:thumbsup:

Application: center garden sized pea

I'll read again in 24 h.

EDIT: Added 2nd computer test

 

[covert24]

wow i know more people received their tubes then this...

Keep up the results guys!!

 

[ICD7]

I was doing some accelerated testing today for a demo and needed to warm up the paste so I threw it in the microwave for 2 min and it was much less viscous.

This is not part of the recommended application procedure but if a couple of you guys want to try it out and give some feedback you are welcome to try.

 

[caferace]

I'm assuming a nice bath in hot water might be less dangerous to the microwave or tube?

-jim

(yes, I understand it isn't conductive, but spewing all the ICD7 into the nuker might be hard to explain to the missus... 🙂)

 

[alfa147x]

Originally posted by: caferace
I'm assuming a nice bath in hot water might be less dangerous to the microwave or tube?

-jim

(yes, I understand it isn't conductive, but spewing all the ICD7 into the nuker might be hard to explain to the missus... 🙂)

thats what I was thinking 😉

 

[ICD7]

I wouldn't do it for 20 min and besides it's not really sealed if it expands the plunger will just move with it. It's mostly bulk with at most a drop or two of of liquid dispersed through it. I might suggest not spreading it with your bare finger...

 

[LOUISSSSS]

so prior to applying, it would be easier if we left the tube in some hot (not boiling) water for about 20 minutes?

my results will be up soon

 

[ICD7]

Not really, a pea sized amount on center, mount heat sink and you are done.

Over the weekends I turn down the heat in my office, 50F when I walk in on a monday the paste flows a little slower on Monday's so I will heat it up. otherwise 20-22C 70F I would not bother. Although I am working on a system application with a large area 38mm by 72mm and as a special case heating up the paste might be a partial solution.

In normal use I would not heat it up

water 4-5 min. 20min water will be at room temp.

 

[LOUISSSSS]

LOUISSSSS' mini-review of ICD7:

Q6600 w/ AS5 running Dual SMP
Core 0 - 52C
Core 1 - 44C
Core 2 - 44C
Core 3 - 50C

Q6600 w/ ICD7 running Dual SMP (after 24 hours of usage)
Core 0 - 53
Core 1 - 47
Core 2 - 47
Core 3 - 51


Don't know idle times, running FAH 24/7 but for the moment during boot up, i saw:
Core 0 - 32C
Core 1 - 29C
Core 2 - 28C
Core 3 - 30C

Notes:
-Temperature readings done with Real Temp 2.0
-Application amount: about 1/4 of the supplied tube, i peeled off the blue sticker.
-Application method: blob in the middle of CPU, attach heatsink directly over it. No spreading, no waiting.

 

[RobertE]

After giving the ICD7 some time to settle in here are my results.

C2D 6320 overclocked to 2.25, temps reported by Speedfan.

### Dynex Silver Compound
Bold ICD7

System...36-38c...35-37c
CPU........50-54c...48-52c
Core 0....39-42c...35-39c
Core 1....37-41c...34-37c

I used a pea sized "drop" if you will. It was very, very thick. After replacing the HS/Fan I checked to see how well it spread out. It didn't spread very much. I was concerned about smoking my CPU so I did used a plastic card to help spread it out. The temps above are with about year old Best Buy/Dynex silver compound and ICD7 after three days.

I do have to say, if I were to buy this, I would be very, very concerned about how thick it is. If I just spend $1000 on the latest & greatest, I'd be very hesitant about cranking up the system hopeing that the thermal paste spreads itself out before damage is done.

 

[ICD7]

Originally posted by: RobertE
After giving the ICD7 some time to settle in here are my results.

C2D 6320 overclocked to 2.25, temps reported by Speedfan.

### Dynex Silver Compound
Bold ICD7

System...36-38c...35-37c
CPU........50-54c...48-52c
Core 0....39-42c...35-39c
Core 1....37-41c...34-37c

I used a pea sized "drop" if you will. It was very, very thick. After replacing the HS/Fan I checked to see how well it spread out. It didn't spread very much. I was concerned about smoking my CPU so I did used a plastic card to help spread it out. The temps above are with about year old Best Buy/Dynex silver compound and ICD7 after three days.

I do have to say, if I were to buy this, I would be very, very concerned about how thick it is. If I just spend $1000 on the latest & greatest, I'd be very hesitant about cranking up the system hopeing that the thermal paste spreads itself out before damage is done.

Thanks for taking the effort to post and test ICD7.

We do get a number of comments on the thickness of the compound and for most it is just that the only experience that they have had is with the wetter style compounds and it is what people are used to

Thousands of tubes out there and have not had any problems to date. The High bulk loadings were a design choice for extended reliability and to minimize any pump out.

Basically it is a lot harder to pump a solid than a liquid, which is the theory behind the OEM type pastes and gels, with reliability taking the priority over performance.

With ICD7 the effort was to maximize performance along with reliability

 

[RobertE]

Forgot to add, the C2D is running Boinc with Einstein 24/7 on both cores.

Overall, happy with the performance so far with the ICD7.

 

[petrusbroder]

Setup for the testing of ICD7:

Main board: _______ MSI MS7250
CPU : ____________ AMD 64 x2 5000+ Black Edition
HSF : ____________ Scythe Katana with supplied fan running at max. speed.
RAM : ____________ 2048 Mbyte, 325 MHz

Room temperature: 24.0 +/- 0.3 ºC (monitored during one week)
Idle: 0% CPU-load
Load: 100% BOINC running POEM@Home
Temps monitored by SpeedFan for 3 hours - log saved
CPU-frequency and Vcore monitored by CPUz
Overclock assumed to stable if no errors were reported by Prime95 for 24 hours.

Arctic Silver applied according to the manufacturer's recommendation.
ICD7 applied with the blob-in-the-middle-and-twist method. The blob was approx 5 mm in diameter. After application and testing I removed the HSF and checked: the compound was evenly spread over the surface of the CPU. Then I cleaned up the HSF and CPU and reapplied ICD7. The cooling results for the OC'd CPU are identical after second application.

Results:

Arctic Silver 5 (cured for 2 weeks at 100% load)
CPU running at stock clock speed and voltage:
Idle: _____ 38 ºC
100% load: 50 ºC
Idle: _____ 38 ºC (after a cool down of one hour)

Overclock: FSB 220 MHz, multiplier x14 speed 3080MHz (from 2600MHz), Vcore 1.289 V
Idle: _____ 39 ºC
100% load: 53 ºC
Idle: _____ 39 ºC (after a cool down of one hour)

ICD7 (cured for 48 hours at 100% load)
CPU running at stock speed and voltage:
Idle: _____ 36 ºC (-3 ºC compared to AS5)
100% load: 46 ºC (-4 ºC compared to AS5)
Idle: _____ 36 ºC (after a cool down of one hour)

Overclock: FSB 240 MHz, multiplier x15 speed 3600MHz (from 2600MHz), Vcore 1.389 V
Idle: _____ 38 ºC
100% load: 53 ºC
Idle: _____ 38 ºC (after a cool down of one hour)

This was so much fun. 1000 MHz overclock is very good. Considering the price of the CPU and the price of ICD7 it is a very good combination. I have now decreased the speed of my CPU to 3300 MHz (FSB 220 MHz, multiplier x15, Vcore: 1.338 V) and it runs rock stable at the following temps:
Idle: _____ 37 ºC
100% load: 50 ºC
Idle: _____ 37 ºC (after a cool down of one hour)

Edit April 11, 2008:
I have had the compound on my computer for 20 additional days. The temps are unchanged compared to the ICD7-numbers above.

IMHO this compound is just great once it has beeen applied. It is not for those who change their CPU/HSF once a week or so, but for us who do not change the CPU or HSF more than once a year or so the stuff is just great.

😀 🙂 😀

 

[Rudy Toody]

My findings on the 10 Second Buzz.

In an earlier post, I had indicated that buzzing the grease did not improve cooling.

That is still the case, however, buzzing is fun!

EDIT Buzzing to spread the grease before setting a tower configuration upright, will create a circular spread about the chip center. When applying heat while upright, the spread will migrate a little to the bottom because of gravity. It might be good to start with a uniform spread. Your mileage may vary.

Steps:

1) place the grease in the center of the chip
2) place the heatsink on the chip and fasten down
3) buzz* for no more that 15 seconds

buzz should be light. I use a barber clipper with the cutters perpendicular to the heatsink/chip. The vibrator I bought is too powerful for 10 Second Buzz, however, it does help my aching back!

The resulting pattern will be somewhere between a US Quarter and US Half-dollar in size. Additional buzzing will not get it to spread any more. During initial burn-in, the grease will spread a little more. These results apply to both Arctic Silver 5 and IC Diamond 7 Carat. By using the buzz, you can get away with using a smaller portion if you wish.

On my initial application of IC Diamond 7 Carat I used about half a pea, buzzed it, overclocked modestly to 14x200 from 13x200. Now I have a full application on the same computer at the same overclock and am getting identical temps

The temps -- using the ZEROtherm Nirvana NV120 Premium CPU Cooler:

49C 46C 47C 29C with room temp 23C
51C 47C 47C 29C with room temp 23C OC to 14x212 --- 2.968GHz! EDIT 2.996GHz was my tops using water cooling.

These correspond to core0, core1, MoBo core temp, MoBo temp.

On of my non-overclocked FX-60s with stock HSF is doing:

58C 57C 56C 29C

Unfortunately, my other box died. Something to do with the BIOS. I have ordered a new one and will get back to you on that.

I am also building a quad-core box which should be fun to buzz.

 

[Insidious]

When I went to that website, mentioned by RudyToody in another post to see prices, I was a little confused by the application instructions... 😱

Application
IC Diamond contains a solvent which enhances application to the CPU. For optimum results, after applying a thin coat of IC Diamond on the CPU, DO NOT install the heatsink for 10 minutes to allow solvent to evaporate.

IC Diamond is composed mostly of diamond powder, and as such is quite thick. Proper application is critical to optimum performance. Squeeze onto the center of the CPU an amount of IC Diamond compound about the size of a pea - the center of the CPU is where most heat is concentrated. Place the heatsink on the CPU and push down to spread IC Diamond over the CPU's surface. Clamp the heatsink and power up the PC.

The first paragraph seems to indicate I need to spread the material in a thin layer and wait for the 10 minutes, then put on the CPU
The second paragraph seems to iindicate I do not need to spread the material....

I know I am just missing something... ICD7, can you help here?

-Sid