HOT ! Various 1TB NVMe with coveted E12 Controller $135 aprox retail

Status
Not open for further replies.

custom90gt

Gawd
Joined
Jun 6, 2007
Messages
547
Since the controller is the only part that should be cooled you could use something like this and use thermal adhesive to fix it in place.

As long as you don't care about a warranty that may work. There still has to be some airflow over the sink because it will just get heatsoaked in a short amount of time. I would go with double sided thermal tape if considering doing something like that so you can remove it if need be.
 

Tricyclthief

Limp Gawd
Joined
Aug 12, 2011
Messages
309
As long as you don't care about a warranty that may work. There still has to be some airflow over the sink because it will just get heatsoaked in a short amount of time. I would go with double sided thermal tape if considering doing something like that so you can remove it if need be.

Thanks for the suggestions, i definitely care about warranty and ended up sending it back. Trying another brand to compare performance.
 

groebuck

2[H]4U
Joined
Mar 9, 2000
Messages
2,531
I ordered a heatsink off of amazon for the inland one - since my 2tb intel will just be doing storage etc. Will see if it helps.
 

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,771
so are we having heat problems because of how incredibly fast this SSD is? Has anyone been doing before and after testing? Results? I'm curious.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
9,259
so are we having heat problems because of how incredibly fast this SSD is? Has anyone been doing before and after testing? Results? I'm curious.

Speed doesnt translate to heat. Voktage and resistance to current does.

Anyways my microcenter 1tb hits about 50 to 60c at peak which is very cool for Nand. Idles at around 35c which is my avg case temp. If you want best nand data retention it should float around 50c all the time. The controller likes cool not hot. The nand likes hot not cool.

Remember what is steaming piping hot to our skin and senses is cold to other things. Like a car motor at op temp is boiling hot and unbearable to people but to a car it's in a sweet comfortable spot.

Honestly if the vendor/manu wanted a heatsink or thought it necessary it would have come with one. Stop getting wrapped around the axle over ssd cooling.
 
Last edited:

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,776
I, for one, won't be spending an additional red cent on this inexpensive component. I will benchmark it after a clean Windows install and enjoy it's state of the art speed until it takes a dirt nap or I upgrade. whichever comes first. :whistle:
 

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,776
Just make note of what oMek ran into w/different chip heights. You wanna cool the controller, not the memory.

So my Alphacool/OCcool heatsink came today, It does not fit. The memory modules on both sides are too tall for the clips. Also noticed the controller is much shorter than them memory.

Im running a Gigabye Aorus Master board, I decided to cut the thermal pad on this stock SSD sink so it only touches the controller and it worked! Running CrystalDisk I hit 45C max. Prior I was hitting 80c+.

That being said, Im still getting about the same benchmarks.

The part you ordered says it is compatible with all single-sided drives, the inland is double sided.
 

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,771
Speed doesnt translate to heat. Voktage and resistance to current does.

Anyways my microcenter 1tb hits about 50 to 60c at peak which is very cool for Nand. Idles at around 35c which is my avg case temp. If you want best nand data retention it should float around 50c all the time. The controller likes cool not hot. The nand likes hot not cool.

Remember what is steaming piping hot to our skin and senses is cold to other things. Like a car motor at op temp is boiling hot and unbearable to people but to a car it's in a sweet comfortable spot.

Honestly if the vendor/manu wanted a heatsink or thought it necessary it would have come with one. Stop getting wrapped around the axle over ssd cooling.

Your logic sounds legit. I came to this conclusion long ago. But, you never know. Some of you guys are hard-core where you leave no stone unturned which I really do appreciate. I was just making sure some new data didn't come to light that I needed to address.
 

Nenu

[H]ardened
Joined
Apr 28, 2007
Messages
19,576
Your logic sounds legit. I came to this conclusion long ago. But, you never know. Some of you guys are hard-core where you leave no stone unturned which I really do appreciate. I was just making sure some new data didn't come to light that I needed to address.
Speed increases current flow of some components proportionally.
His statement is wrong.

If you increase a GPUs speed (or any processor) 10% it will use approx 10% more power.
 

Joust

Supreme [H]ardness
Joined
Nov 30, 2017
Messages
4,243
Speed increases current flow of some components proportionally.
His statement is wrong.

If you increase a GPUs speed (or any processor) 10% it will use approx 10% more power.

Nah. Plenty doesn't scale in a linear fashion. Efficiencies drop.
 

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,771
I mean, if I have to I can apply a little tiny copper heatsink to the controller on this SSD but I bet that would be a waste of time.

Regardless I love that you guys are eating this little gem of a 1tb NVMe SSD drive up.

Pretty 'f'king amazing for the performance and value. Cannot be beat .... period.
 

Tricyclthief

Limp Gawd
Joined
Aug 12, 2011
Messages
309
so are we having heat problems because of how incredibly fast this SSD is? Has anyone been doing before and after testing? Results? I'm curious.

Not sure about the others, but my heat is either from a crap drive or my computer/mobo layout. Either way i sent it back, hitting almost 80c is just nuts. Have a different brand phison that will be here Tuesday, i'll find out if it's the controller or was just the Inland drive.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
9,259
Speed increases current flow of some components proportionally.
His statement is wrong.

If you increase a GPUs speed (or any processor) 10% it will use approx 10% more power.

Speed is just a measure of what we want to qualitatively observe. It has nothing to do with physics. Speed something up slow it down... blag blah.

We have physics formulas that answer all of the above.

Voltage, resistance, and current are what matters. You can increase the "speed" of an object and increase its efficiency in materials and design and actually decrease heat generated. I.e. 14nm to 7nm generates less heat at the same or greater frequencies aka "speed".

A 5 speed transmission might allow a car to go 80mph but at 3500rpm. A 6 speed transmission might allow same car to go 80mph at 2200 rpm. Its rotating less at the engine and consuming less fuel and still generating the same kw/hp output necessary to go 80mph.

It's about design efficiency. Not speed.

Since we cant overclock the SSD (normally) it's running a fixed parameter as far as thermals so its "speed" has no bearing on its thermal generation. That's a function of the electrical design of the actual ssd. If we were to OC the ssd somehow we would pump more voltage through a fixed and already designed substrate so current would increase but resistance would still impede the flow of electrons thus elevating the thermal output. This is because in this scenario we are forcing more electrons through the same space than it was designed. Increased resistance and more electromotive force equals more heat. I'm trying to explain this as layman's as possible.
 
Last edited:

Nenu

[H]ardened
Joined
Apr 28, 2007
Messages
19,576
Speed is just a measure of what we want to qualitatively observe. It has nothing to do with physics. Speed something up slow it down... blag blah.

We have physics formulas that answer all of the above.

Voltage, resistance, and current are what matters. You can increase the "speed" of an object and increase its efficiency in materials and design and actually decrease heat generated. I.e. 14nm to 7nm generates less heat at the same or greater frequencies aka "speed".

A 5 speed transmission might allow a car to go 80mph but at 3500rpm. A 6 speed transmission might allow same car to go 80mph at 2200 rpm. Its rotating less at the engine and consuming less fuel and still generating the same kw/hp output necessary to go 80mph.

It's about design efficiency. Not speed.

Since we cant overclock the SSD (normally) it's running a fixed parameter as far as thermals so its "speed" has no bearing on its thermal generation. That's a function of the electrical design of the actual ssd. If we were to OC the ssd somehow we would pump more voltage through a fixed and already designed substrate so current would increase but resistance would still impede the flow of electrons thus elevating the thermal output. I'm trying to explain this as layman's as possible.
When you increase the speed of a digital device you reduce its effective resistance, more current flows proportionally.
Not sure what you are on about tbh.
 

MyNameIsAlex

Limp Gawd
Joined
Mar 10, 2019
Messages
313
When you increase the speed of a digital device you reduce its effective resistance, more current flows proportionally.
Not sure what you are on about tbh.

He is saying faster speed with no voltage change will not affect temps?

I don’t understand that people that think they will gain performance if they cool the drive down. All you will do increase lifespan
 

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,776
I don’t understand that people that think they will gain performance if they cool the drive down. All you will do increase lifespan
The concern is that the controller is throttling due to temp which in turn would slow it down, so in that case proper cooling would indeed cause a performance increase.
 

Nenu

[H]ardened
Joined
Apr 28, 2007
Messages
19,576
He is saying faster speed with no voltage change will not affect temps?

I don’t understand that people that think they will gain performance if they cool the drive down. All you will do increase lifespan
The problem is the drive will throttle (reduce its speed to reduce power consumption) once it reaches a certain temp.
Cooling the drive will increase performance by reducing the amount of throttling.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
9,259
When you increase the speed of a digital device you reduce its effective resistance, more current flows proportionally.
Not sure what you are on about tbh.

I have no idea what you are trying to state by saying, increasing speed equals more heat?

Then how the hell do you overclock an SSD? You gonna hack the firmware and increase the clock speed of the controller? Then by all means have at it.

Going from last generation to a next generation increases speed but it doesnt increase heat (in almost all cases) because new efficiencies in the design and silicon etc... lower thermal output and increase throughput.

So your arguing at this point to save face and nothing more unless you have something to add that explains why an increase in throughput from last gen Inland drives vs. this generation inland drives translates to more heat? We went from 2x pcie lanes to 4x pcie lanes and a brand new controller design was implemented? Temperatures remain the same from the old gen to the new gen. Yet the throughput has increased by factors more.
 

Nenu

[H]ardened
Joined
Apr 28, 2007
Messages
19,576
I have no idea what you are trying to state by saying, increasing speed equals more heat?
Basic facts for the same silicon.

Then how the hell do you overclock an SSD? You gonna hack the firmware and increase the clock speed of the controller? Then by all means have at it.
Not sure what you are on about.

Going from last generation to a next generation increases speed but it doesnt increase heat (in almost all cases) because new efficiencies in the design and silicon etc... lower thermal output and increase throughput.
Many different methods reduce heat/power consumption.
Die shrinks primarily.

So your arguing at this point to save face and nothing more unless you have something to add that explains why an increase in throughput from last gen Inland drives vs. this generation inland drives translates to more heat? We went from 2x pcie lanes to 4x pcie lanes and a brand new controller design was implemented? Temperatures remain the same from the old gen to the new gen. Yet the throughput has increased by factors more.
Make up what you like.
 

MyNameIsAlex

Limp Gawd
Joined
Mar 10, 2019
Messages
313
The concern is that the controller is throttling due to temp which in turn would slow it down, so in that case proper cooling would indeed cause a performance increase.

The problem is the drive will throttle (reduce its speed to reduce power consumption) once it reaches a certain temp.
Cooling the drive will increase performance by reducing the amount of throttling.

Do we have any evidence to support these concerns?
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
9,259
Basic facts for the same silicon.


Not sure what you are on about.


Many different methods reduce heat/power consumption.
Die shrinks primarily.


Make up what you like.

I wasnt attacking you by suggesting saving face... I do it all the time, save face, its the American, Western World way man!


If you want the skinny on it ...

The Phison E8 in the old 1TB uses about 3 watts at max load @ 3.3v

The Phison E12 in the new 1tb uses about 7.5 watts at max load @ 3.3v

yet the operating temperatures listed on their site is the same exact 0 - 70c.

If you own both, and I do now, the op temps are so insanely close to one another I am not sure how these numbers even translate to more heat or less heat. Its funny (not you) but the whole numbers game they are posting vs. real world.


The jest of what I have been saying is, who the F cares. If you can actually hit 70c and hold it there then your using the drive like its meant to be used. Toss a heatsink on it if you like. Most motherboards have built in nvme heatsinks already. The only functional heatsink I have seen is probably the one of the WD 750 Black Edition probably because I understand it is made by EK(?). Unless your going to be storing data on your SSD for the next 96 years maybe getting the perfect op temp is suggestable, but by the time even 1% of the NAND lifespan is used, I bet 99.9999% of SSD owners would have changed their drive twice over by that time. I have a 960 pro I have written the library of congress worth of data probably 20x and it just now hit 99% life span remaining.

Here they are snake oil:
https://www.techpowerup.com/230293/msi-m-2-shield-is-snake-oil-say-tests-company-refutes-charges

But Tweaktown swears they make a difference:
https://www.tweaktown.com/articles/8599/investigation-heatsinks-worth/index.html

Hey if you get a heatsink just blow air on it to make it work right I guess.
 
Last edited:

Nenu

[H]ardened
Joined
Apr 28, 2007
Messages
19,576
If you want the skinny on it ...

The Phison E8 in the old 1TB uses about 3 watts at max load @ 3.3v

The Phison E12 in the new 1tb uses about 7.5 watts at max load @ 3.3v

yet the operating temperatures listed on their site is the same exact 0 - 70c.

If you own both, and I do now, the op temps are so insanely close to one another I am not sure how these numbers even translate to more heat or less heat. Its funny (not you) but the whole numbers game they are posting vs. real world.


The jest of what I have been saying is, who the F cares. If you can actually hit 70c and hold it there then your using the drive like its meant to be used. Toss a heatsink on it if you like. Snake oil man. Most motherboards have built in nvme heatsinks already. The only functional heatsink I have seen is probably the one of the WD 750 Black Edition. Even the EK heatsink I have on my old 960 pro made zilch difference in the op temps. They are snake oil man.
Operating temps on different silicon are not proportional for the same speed.
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
9,259
It’s gist, not jest..... ;)

you are correct!

Nenu I recommend you try this drive. you will love it regardless of the power to heat to blah blah! Its a damn spiffy and quick drive.

better than this old piece of s&&t 128GB Inland nVME on this 2600x im typing all these useless replies on:
upload_2019-3-24_22-26-18.png


here is the new 1tb with only 200'ish GB free left on the drive (I have been slamming it all day with h265 conversions non stop).

upload_2019-3-24_22-28-14.png
 
Last edited:

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,771
The concern is that the controller is throttling due to temp which in turn would slow it down, so in that case proper cooling would indeed cause a performance increase.

I think this is at least worth while for a few of you to take to task and report back.

I would also be careful with reported motherboard temps on the SSD, I think I've read this number is often not correct? I'm seeing numbers / temps from 45c idle to 58c in-use.

Maybe a heat probe. But would be interesting to get some numbers and facts back into this tread just so everyone is up to date with these numbers.
 

Hashiriya415

Limp Gawd
Joined
Mar 17, 2019
Messages
128
I just bought a 660p. But now that this is just $35 more. Should I return my 660p for this instead? Mostly gaming, light video editing
 

troisanh

Supreme [H]ardness
Joined
Dec 27, 2005
Messages
7,837
anyone got spare 1T drive they wanna sell PST. 1TB nvme m.2 premium to be exact.

no microcenter nearby and can not order online anymore.
 
Status
Not open for further replies.
Top