"...a graphene copper foil heat dissipation module that is only 1 mm thick."
The marketing statement of the century. Someone give the clowns at Team Group a cookie for inventing a new flavor of snake oil. I'm sure the 30 something Lego-loving basement dwelling man children running RGB-festooned "battle stations" off mommy's power outlets will drink it up.
IHS of the ssd. same shit, same shitty result except it might help to dissipate if we have ssd shield? Graphene was incredible at spreading heat around. Can I get a test of this ssd with and without the foil under same mobo & ssd shield?
I really thought that would be a pretty specific, limited subset of people. I must report with some disappointment that I was wrong and the brush strokes were indeed wide.
...
Except for shabby. Mooching is great for some people. I married a moocher at one point and put up with mooching for close to twenty years. Now some unsuspecting woman in another nation is supporting the ex-moocher that never grew up. I hope those Lego skills land a great paying job someday so I can finally get a child support check or two out of that deadbeat, but it's been a few years already and I suspect that minimum wage is optimistic at best.
Was that who you were describing in your original post, your ex husband? I'm going to say yes since the correlation between graphene copper foil and Lego-loving basement dwelling man child is pretty slim.
Thank you. Yes, my inarticulate, bitter rampage about the computer industry and the suckers that get preyed upon by "clever" sales techniques that anyone with a couple of business college courses would see through was intended to point that out.
It's just a sticker. It doesn't have significant mass or substantially increase the surface area which are both helpful in dissipating heat generated by electronic components. It could be made of rainbows and Harry Potter's dandruff OR graphene (apparently the newest marketing magic product differentiator) and copper and it would still do the same thing which is merely serve as a place on which to display the company's logo and name. Swinging a sticker as thermally beneficial is flirting with the line between sales speak and false advertising. The legal department's lawyer(s) must be sleeping in the wheelhouse if that claim is actually out of Team Group's product advertising.
If I were advising them, I would have pointed them at any number of class action lawsuits that demonstrate the considerable risk the company is accepting at this point. Basement dwelling adult moochers aside, there are some fairly sophisticated consumers that purchase individual components. They could run convincing tests that will most likely debunk the thermal performance claims being made. It would have been much safer (although still shady as f*ck) to simply state the materials in the sticker and not proclaim a benefit. Buyers already falsely convinced that graphene = magic cooling would make the assumption on their own that the sticker is beneficial and Team Group would accept no increased risk of legal entanglement. Doing what they're doing now is just flipping stupid.
Now I would love to see your reaction if this actually gets reviewed and verified to do what it is advertised for. Having strong words about something without any evidence to support it is a bit stupid tbh. Though I agree with your expectation that the magic sticker won't turn to be a magic cooler.
Hopefully Anandtech gets access to one and has some other sticker available to slap on it for testing. I'd like AT to apply an aluminum sticker as a comparative measure in particular. It would be great to see if Team Group's claims can be supported in a third party run test under controlled conditions. If that sort of test does yield meaningful results in favor of the sticker than I'll happily retract my statements regarding doubt about the sticker's proclaimed performance benefits.
There is no need to be snide and belittle others. This is especially true when some of your statements and understanding are incorrect. For example mass does not play a roll in heat transfer. You will not find mass in equations for conductive, convective, or black body radiation heat transfer. The article states, "Meanwhile, according to Team Group, the graphene copper foil heat dissipation sticker reduces controller’s temperature from 61.8°C to 56.5°C when compared to the lack of any heat spreader." A 5.3°C decrease seems reasonable. These controllers generate far less heat than a CPU or GPU. Graphen's conductive heat transfer coefficient is an order of magnitude greater than copper. Team Group does not disclose how much of the 1mm thick sticker is graphene and how much is copper, but in the worst case the sticker is all copper. A 1mm X 20mm copper sheet should provide sufficient cross-sectional area to move the heat way from this type of controller. Heat transfer then is convective limited. The surface area of heat spreader is larger than the controller since it covers both the controller and NAND flash dies. This greater surface area improves convective heat transfer. Which means the heat spreader or sticker will lower the temperature of the controller compared the controller without the spreader. The proof is in the pudding. If Team Group's testing shows a 5.1°C drop in between an SSD with and without the spreader, given the same environment, than their heat spreader improves cooling while staying within the form factor limits imposed by laptops. If the test were performed in a laptop with internal air temperature between 30°C to 50°C and airflow near around the SSD is limited, as is the case with my Gigabyte Aero 15x, than the heat spreader is effective. If the test was performed at 20°C to 25°C in a desktop with good airflow than the heat spreader is more of a gimmick.
You may want to do a little research ahead of time. The rate of heat transfer is not the only factor to consider. Isolating that alone to justify quoting the article without consideration for the testing conditions or even the limited disclaimer coverage Team Group attempted to build into the statemetn that did end up in the article doesn't exactly help your case. Presumptions about the effectiveness of graphene and how its used in this circumstance ignores implementation in favor of simply believing a certain material tossed in the ingredients list will impart all of the properties of that material automatically.
On a more upbeat note, you might be part of the target market for this SSD. Buy one and enjoy it if that's your thing. Feel absolutely free to presume the sticker is making some sort of functioanl difference while it's in your system.
Only one who made blind assumptions about use cases was you. The result was in the article, a delta of 5.1 C is significant under identical test conditions. You repeatedly show off your ignorance on subjects outside your scope and then counter-punch when someone schools you. It would probably be in your own best interest to check Wikipedia or something before spouting off about a WELL understood area of physics/engineering like heat transfer.
"You will not find mass in equations for conductive, convective, or black body radiation heat transfer."
That is only correct if you look at the steady state heat transfer exclusively, which is not realistic for the application at hand. Here we will have bursts of heat production in many real-world use cases, and the absolute heat capacity of the sticker (in J/K) will matter, and that is specific heat capacity * mass. A low-weight sticker can't smooth out temperature peaks as well as a larger heatsink. Obviously, whether steady-state or dynamic heat transfer are more important will depend on the application and environment, but simply ignoring the dynamic properties of the system seems clearly flawed.
So I would argue that your statements and understanding in this matter are incorrect as well.
"It doesn't have significant mass or substantially increase the surface area which are both helpful in dissipating heat "
1. Totally false about the mass - heat transfer resistance is proportional to thickness, higher mass comes with higher thickness. Physics (Fourier’s law of heat conduction).
2. Area compared to what? That only chip which heats up significantly - the controller? The sticker is quite a bit bigger.
Your question is front-loaded. You imply that my criticisms of the sticker is also an attack on the physical properties of some the materials that comprise it in order to preposition a follow up argument when I reply. This isn't my first rodeo, little one so you may as well hang up that tactic at the front door now that you're walking into my proverbial bar.
With that out of the way, did you have something constructive to discuss about your new Lego Star Destroyer or are we done here?
Drinking booze is a huge waste of money. I'd rather benefit from compound interest. Also it sets a bad example for children to even keep stupid vices like that stocked in a home.
I love how this comment thread revolves around one comment. Also just wanted to mention, graphene may be an awesome conductor of heat, but air is a terrible conductor of heat. There's no way you can get all that heat of a mm thin piece of graphene cuz the air won't just heat up and remove the heat quick enough. This is why high end air coolers have so many fins, to maximize the amount of AIR that can come in contact with the heat spreader and TAKE AWAY the heat.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
28 Comments
Back to Article
PeachNCream - Wednesday, October 9, 2019 - link
"...a graphene copper foil heat dissipation module that is only 1 mm thick."The marketing statement of the century. Someone give the clowns at Team Group a cookie for inventing a new flavor of snake oil. I'm sure the 30 something Lego-loving basement dwelling man children running RGB-festooned "battle stations" off mommy's power outlets will drink it up.
The True Morbus - Wednesday, October 9, 2019 - link
What's wrong with liking Lego?damianrobertjones - Wednesday, October 9, 2019 - link
What's wrong with liking Lego? x2(I own my own house, with no basement)
deil - Wednesday, October 9, 2019 - link
IHS of the ssd. same shit, same shitty result except it might help to dissipate if we have ssd shield?Graphene was incredible at spreading heat around. Can I get a test of this ssd with and without the foil under same mobo & ssd shield?
shabby - Wednesday, October 9, 2019 - link
What's wrong with being moocher?29a - Wednesday, October 9, 2019 - link
You paint with a rather large wide brush.PeachNCream - Wednesday, October 9, 2019 - link
I really thought that would be a pretty specific, limited subset of people. I must report with some disappointment that I was wrong and the brush strokes were indeed wide....
Except for shabby. Mooching is great for some people. I married a moocher at one point and put up with mooching for close to twenty years. Now some unsuspecting woman in another nation is supporting the ex-moocher that never grew up. I hope those Lego skills land a great paying job someday so I can finally get a child support check or two out of that deadbeat, but it's been a few years already and I suspect that minimum wage is optimistic at best.
shabby - Wednesday, October 9, 2019 - link
Was that who you were describing in your original post, your ex husband? I'm going to say yes since the correlation between graphene copper foil and Lego-loving basement dwelling man child is pretty slim.rahvin - Wednesday, October 9, 2019 - link
It's a sticker, not a heat spreader. So they put a little copper foil in their sticker, that doesn't make it a realPeachNCream - Wednesday, October 9, 2019 - link
Thank you. Yes, my inarticulate, bitter rampage about the computer industry and the suckers that get preyed upon by "clever" sales techniques that anyone with a couple of business college courses would see through was intended to point that out.It's just a sticker. It doesn't have significant mass or substantially increase the surface area which are both helpful in dissipating heat generated by electronic components. It could be made of rainbows and Harry Potter's dandruff OR graphene (apparently the newest marketing magic product differentiator) and copper and it would still do the same thing which is merely serve as a place on which to display the company's logo and name. Swinging a sticker as thermally beneficial is flirting with the line between sales speak and false advertising. The legal department's lawyer(s) must be sleeping in the wheelhouse if that claim is actually out of Team Group's product advertising.
If I were advising them, I would have pointed them at any number of class action lawsuits that demonstrate the considerable risk the company is accepting at this point. Basement dwelling adult moochers aside, there are some fairly sophisticated consumers that purchase individual components. They could run convincing tests that will most likely debunk the thermal performance claims being made. It would have been much safer (although still shady as f*ck) to simply state the materials in the sticker and not proclaim a benefit. Buyers already falsely convinced that graphene = magic cooling would make the assumption on their own that the sticker is beneficial and Team Group would accept no increased risk of legal entanglement. Doing what they're doing now is just flipping stupid.
HollyDOL - Thursday, October 10, 2019 - link
Now I would love to see your reaction if this actually gets reviewed and verified to do what it is advertised for.Having strong words about something without any evidence to support it is a bit stupid tbh.
Though I agree with your expectation that the magic sticker won't turn to be a magic cooler.
PeachNCream - Thursday, October 10, 2019 - link
Hopefully Anandtech gets access to one and has some other sticker available to slap on it for testing. I'd like AT to apply an aluminum sticker as a comparative measure in particular. It would be great to see if Team Group's claims can be supported in a third party run test under controlled conditions. If that sort of test does yield meaningful results in favor of the sticker than I'll happily retract my statements regarding doubt about the sticker's proclaimed performance benefits.bugnguts - Thursday, October 10, 2019 - link
There is no need to be snide and belittle others. This is especially true when some of your statements and understanding are incorrect. For example mass does not play a roll in heat transfer. You will not find mass in equations for conductive, convective, or black body radiation heat transfer.The article states, "Meanwhile, according to Team Group, the graphene copper foil heat dissipation sticker reduces controller’s temperature from 61.8°C to 56.5°C when compared to the lack of any heat spreader." A 5.3°C decrease seems reasonable. These controllers generate far less heat than a CPU or GPU.
Graphen's conductive heat transfer coefficient is an order of magnitude greater than copper. Team Group does not disclose how much of the 1mm thick sticker is graphene and how much is copper, but in the worst case the sticker is all copper. A 1mm X 20mm copper sheet should provide sufficient cross-sectional area to move the heat way from this type of controller. Heat transfer then is convective limited.
The surface area of heat spreader is larger than the controller since it covers both the controller and NAND flash dies. This greater surface area improves convective heat transfer. Which means the heat spreader or sticker will lower the temperature of the controller compared the controller without the spreader.
The proof is in the pudding. If Team Group's testing shows a 5.1°C drop in between an SSD with and without the spreader, given the same environment, than their heat spreader improves cooling while staying within the form factor limits imposed by laptops. If the test were performed in a laptop with internal air temperature between 30°C to 50°C and airflow near around the SSD is limited, as is the case with my Gigabyte Aero 15x, than the heat spreader is effective. If the test was performed at 20°C to 25°C in a desktop with good airflow than the heat spreader is more of a gimmick.
PeachNCream - Thursday, October 10, 2019 - link
You may want to do a little research ahead of time. The rate of heat transfer is not the only factor to consider. Isolating that alone to justify quoting the article without consideration for the testing conditions or even the limited disclaimer coverage Team Group attempted to build into the statemetn that did end up in the article doesn't exactly help your case. Presumptions about the effectiveness of graphene and how its used in this circumstance ignores implementation in favor of simply believing a certain material tossed in the ingredients list will impart all of the properties of that material automatically.On a more upbeat note, you might be part of the target market for this SSD. Buy one and enjoy it if that's your thing. Feel absolutely free to presume the sticker is making some sort of functioanl difference while it's in your system.
FullmetalTitan - Thursday, October 10, 2019 - link
Only one who made blind assumptions about use cases was you. The result was in the article, a delta of 5.1 C is significant under identical test conditions. You repeatedly show off your ignorance on subjects outside your scope and then counter-punch when someone schools you. It would probably be in your own best interest to check Wikipedia or something before spouting off about a WELL understood area of physics/engineering like heat transfer.PeachNCream - Thursday, October 10, 2019 - link
Lego fans unite!JanW1 - Friday, October 11, 2019 - link
"You will not find mass in equations for conductive, convective, or black body radiation heat transfer."That is only correct if you look at the steady state heat transfer exclusively, which is not realistic for the application at hand. Here we will have bursts of heat production in many real-world use cases, and the absolute heat capacity of the sticker (in J/K) will matter, and that is specific heat capacity * mass. A low-weight sticker can't smooth out temperature peaks as well as a larger heatsink. Obviously, whether steady-state or dynamic heat transfer are more important will depend on the application and environment, but simply ignoring the dynamic properties of the system seems clearly flawed.
So I would argue that your statements and understanding in this matter are incorrect as well.
peevee - Thursday, October 10, 2019 - link
"It doesn't have significant mass or substantially increase the surface area which are both helpful in dissipating heat "1. Totally false about the mass - heat transfer resistance is proportional to thickness, higher mass comes with higher thickness. Physics (Fourier’s law of heat conduction).
2. Area compared to what? That only chip which heats up significantly - the controller? The sticker is quite a bit bigger.
PeachNCream - Thursday, October 10, 2019 - link
Why bother with mass or fins at all then? Why do processor heat sinks have all that weight and size?NGneer - Wednesday, October 9, 2019 - link
Just curious what you technically don't like about a graphene-copper composite for heat spreading?PeachNCream - Wednesday, October 9, 2019 - link
Your question is front-loaded. You imply that my criticisms of the sticker is also an attack on the physical properties of some the materials that comprise it in order to preposition a follow up argument when I reply. This isn't my first rodeo, little one so you may as well hang up that tactic at the front door now that you're walking into my proverbial bar.With that out of the way, did you have something constructive to discuss about your new Lego Star Destroyer or are we done here?
hbsource - Thursday, October 10, 2019 - link
........ Why you shouldn't drink before using the internet.peevee - Thursday, October 10, 2019 - link
How one survives it otherwise? ;)PeachNCream - Thursday, October 10, 2019 - link
Drinking booze is a huge waste of money. I'd rather benefit from compound interest. Also it sets a bad example for children to even keep stupid vices like that stocked in a home.8lec - Tuesday, October 15, 2019 - link
I love how this comment thread revolves around one comment.Also just wanted to mention, graphene may be an awesome conductor of heat, but air is a terrible conductor of heat. There's no way you can get all that heat of a mm thin piece of graphene cuz the air won't just heat up and remove the heat quick enough. This is why high end air coolers have so many fins, to maximize the amount of AIR that can come in contact with the heat spreader and TAKE AWAY the heat.
8lec - Wednesday, October 16, 2019 - link
Feel free to correct me if I'm wrongmercury12288 - Saturday, January 11, 2020 - link
I have a Gen4 PCIe4 system (ROG Zenith II) here is performance of this SSD:$ sudo fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=test --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75
test: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
fio-3.1
Starting 1 process
test: Laying out IO file (1 file / 4096MiB)
Jobs: 1 (f=1)
test: (groupid=0, jobs=1): err= 0: pid=9868: Sat Jan 11 17:58:49 2020
read: IOPS=302k, BW=1181MiB/s (1238MB/s)(3070MiB/2600msec)
bw ( MiB/s): min= 1050, max= 1304, per=100.00%, avg=1182.76, stdev=123.43, samples=5
iops : min=269036, max=333994, avg=302786.20, stdev=31597.57, samples=5
write: IOPS=101k, BW=395MiB/s (414MB/s)(1026MiB/2600msec)
bw ( KiB/s): min=361528, max=444736, per=100.00%, avg=405068.00, stdev=40772.91, samples=5
iops : min=90382, max=111184, avg=101267.00, stdev=10193.23, samples=5
cpu : usr=20.97%, sys=76.88%, ctx=10505, majf=0, minf=9
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0%
issued rwt: total=785920,262656,0, short=0,0,0, dropped=0,0,0
latency : target=0, window=0, percentile=100.00%, depth=64
Run status group 0 (all jobs):
READ: bw=1181MiB/s (1238MB/s), 1181MiB/s-1181MiB/s (1238MB/s-1238MB/s), io=3070MiB (3219MB), run=2600-2600msec
WRITE: bw=395MiB/s (414MB/s), 395MiB/s-395MiB/s (414MB/s-414MB/s), io=1026MiB (1076MB), run=2600-2600msec
Disk stats (read/write):
nvme0n1: ios=785392/262517, merge=0/0, ticks=102676/3481, in_queue=4, util=96.47%
mercury12288 - Saturday, January 11, 2020 - link
Big Test:$ sudo fio --name=rand --ioengine=libaio --iodepth=1 --rw=randrw --bs=4m --direct=0 --size=40G --numjobs=4 --runtime=240 --group_reporting
randwrite: (g=0): rw=randrw, bs=(R) 4096KiB-4096KiB, (W) 4096KiB-4096KiB, (T) 4096KiB-4096KiB, ioengine=libaio, iodepth=1
...
fio-3.1
Starting 4 processes
randwrite: Laying out IO file (1 file / 40960MiB)
randwrite: Laying out IO file (1 file / 40960MiB)
randwrite: Laying out IO file (1 file / 40960MiB)
randwrite: Laying out IO file (1 file / 40960MiB)
Jobs: 4 (f=4): [m(4)][97.7%][r=1952MiB/s,w=1972MiB/s][r=488,w=493 IOPS][eta 00m:01s]
randwrite: (groupid=0, jobs=4): err= 0: pid=10259: Sat Jan 11 18:25:00 2020
read: IOPS=472, BW=1890MiB/s (1982MB/s)(79.2GiB/42891msec)
slat (usec): min=1778, max=508884, avg=6355.06, stdev=7529.55
clat (nsec): min=360, max=6372, avg=761.29, stdev=204.92
lat (usec): min=1779, max=508887, avg=6356.40, stdev=7529.65
clat percentiles (nsec):
| 1.00th=[ 402], 5.00th=[ 442], 10.00th=[ 502], 20.00th=[ 644],
| 30.00th=[ 684], 40.00th=[ 708], 50.00th=[ 740], 60.00th=[ 780],
| 70.00th=[ 836], 80.00th=[ 900], 90.00th=[ 980], 95.00th=[ 1048],
| 99.00th=[ 1208], 99.50th=[ 1400], 99.90th=[ 2288], 99.95th=[ 2992],
| 99.99th=[ 5536]
bw ( KiB/s): min=172032, max=892928, per=25.17%, avg=487300.45, stdev=102241.30, samples=337
iops : min= 42, max= 218, avg=118.91, stdev=24.97, samples=337
write: IOPS=482, BW=1930MiB/s (2023MB/s)(80.8GiB/42891msec)
slat (usec): min=889, max=44128, avg=1970.38, stdev=636.91
clat (nsec): min=331, max=11721, avg=644.75, stdev=165.49
lat (usec): min=889, max=44130, avg=1971.41, stdev=636.97
clat percentiles (nsec):
| 1.00th=[ 370], 5.00th=[ 390], 10.00th=[ 422], 20.00th=[ 572],
| 30.00th=[ 604], 40.00th=[ 620], 50.00th=[ 628], 60.00th=[ 652],
| 70.00th=[ 700], 80.00th=[ 772], 90.00th=[ 820], 95.00th=[ 868],
| 99.00th=[ 964], 99.50th=[ 1004], 99.90th=[ 1144], 99.95th=[ 1560],
| 99.99th=[ 3856]
bw ( KiB/s): min=180224, max=966656, per=25.19%, avg=497651.17, stdev=120205.22, samples=337
iops : min= 44, max= 236, avg=121.44, stdev=29.35, samples=337
lat (nsec) : 500=13.07%, 750=49.88%, 1000=32.57%
lat (usec) : 2=4.38%, 4=0.10%, 10=0.01%, 20=0.01%
cpu : usr=0.72%, sys=35.32%, ctx=84445, majf=0, minf=49
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwt: total=20270,20690,0, short=0,0,0, dropped=0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=1890MiB/s (1982MB/s), 1890MiB/s-1890MiB/s (1982MB/s-1982MB/s), io=79.2GiB (85.0GB), run=42891-42891msec
WRITE: bw=1930MiB/s (2023MB/s), 1930MiB/s-1930MiB/s (2023MB/s-2023MB/s), io=80.8GiB (86.8GB), run=42891-42891msec
Disk stats (read/write):
nvme0n1: ios=82657/82416, merge=0/184, ticks=112137/2581586, in_queue=2429296, util=99.46%