Production FormD T1 Classic (READ FIRST POST)

[gwertheim]

You'd be surprised, here are a few sources:
EKWB
LTT forum
A reddit thread

As you can see, a 100 degrees Celsius isn't exactly nothing. You'd definitely want bigger radiators, pumps and fans in general to use with a monoblock.
Especially since with every next CPU, the voltages and TDPs increase. As a result, the VRMs will work in overdrive to keep up with the CPU's demands. Some mobo manufacturers even add to their VRMs their own dedicated mini fan lately, if that makes sense.

Example:

1. LTT IS A HACK

2. Unless they state their Education, THEY DON'T KNOW ANYTHING. Referring to them is like referring to a Wikipedia article without footnotes.

3. Refer to #1

 

[Lallo]

What? Are the FormD-T1 out of stock now? I thought you would be able to order and it would ship in December....

 

[Arc_Lag]

Well watts of waste heat and temperature are 2 very different things. You can add 10 watts of heat to a small block of metal and get it very hot. To maintain temperature,the cooling solution need only remove 10 watts of heat in that scenario. If the cooling solution only removes 5 watts? Well then your temperatures will climb.

This is how much heat a VRM gives off:
Don't trust a random guy on reddit? Here is a buildzoid video (jump to ~10mins)
Again, temperature is a measure of energy at a given time, a snapshot if you will. Watts is a unit of power which is energy over time. You absolutely cannot compare the two. It's like saying "look at how much water is in this glass, it must've been filled up very fast!" when it very well could be a slow trickle over a long time.

I think the real reason for the high CPU temps is poor contact/pressure. Either the monoblock was machined wrong or the thermal pads should be thinner or both.

Thank you for elucidating!

 

[Allhopeforhumanity]

Well watts of waste heat and temperature are 2 very different things

Until you venture into quantum and start looking at the vibrations of a single molecule... I'll see myself out.

 

[robojim]

Yeah, interesting thought since Ampere ought to be relatively well distributed in the market by then. Although I also wonder whether it's a calculated risk based on possible assumptions that hardcore NVIDIA fans probably wouldn't be swayed anyhow, and those on the fence are likely already awaiting Big Navi, et al. anyhow. Still seems hella risky to me to wait that long, though. Hopefully it's worth it!

Ampere cards' supply is gonna be weak, especially for 3090 but even 3080 probably won't catch up till later in the year according to leaks. I don't think AMD is too worried. Also, I don't think AMD is planning to recapture marketshare heavily either.

 

[Madhawk1995]

https://noctua.at/en/nf-a12x15-pwm
vs
https://noctua.at/en/nf-a12x25-pwm/specification
airflow 94 vs 102 m³/h
quiet small difference !!
why you even bother with thicker fan.

.

We are throwing these on radiators. So, compare the static pressure.
15mm: 1,53 mm H₂O
25mm: 2,34 mm H₂O.
Also, I prefer the sound profile of the nf-a12x25 more than the nf-a12x15. Only one of them at 1500 rpm annoys me. Guess which one?

 

[SoSquidTaste]

Guess which one?

NF-A4x10?!?

Relevance edit: Yeah, imagine my shock on first boot when my Quadro had all three of my A12x15 fans at full blast. I was about to nope the hell out of my build until I confirmed that they were all at full blast. There's a symphony...

 

[eedev]

We are throwing these on radiators. So, compare the static pressure.
15mm: 1,53 mm H₂O
25mm: 2,34 mm H₂O.
Also, I prefer the sound profile of the nf-a12x25 more than the nf-a12x15. Only one of them at 1500 rpm annoys me. Guess which one?

The nf-a12x15-pwm is really noticeable.

At first it was touching my motherboard (Z490I AORUS ULTRA).
I had to use 2-3 layers of silentblocs as spacers and not screw too much for the noise to be reduced but it's not enough for my taste.

I think I'm going to use a specific cable for this one and lower the rpm.

 

[thelaughingman]

We are throwing these on radiators. So, compare the static pressure.
15mm: 1,53 mm H₂O
25mm: 2,34 mm H₂O.
Also, I prefer the sound profile of the nf-a12x25 more than the nf-a12x15. Only one of them at 1500 rpm annoys me. Guess which one?

I prefer the A12x25 too but unless we sand them down to use with the TX240, it won't fit a watercooled T1 ?

 

[stanleyguan]

I prefer the A12x25 too but unless we sand them down to use with the TX240, it won't fit a watercooled T1 ?

Takers?

The only thing preventing me from getting them and sanding them down is the fact that they are brown. Supposedly the black ones are coming in Q1.

 

[thelaughingman]

Takers?

The only thing preventing me from getting them and sanding them down is the fact that they are brown. Supposedly the black ones are coming in Q1.

I'm seriously considering doing it for the browns since I have 3 in my builds right now. The 2 that are currently on the GPU will become redundant once I get RDNA2 and put it on water anyway.
@DrHudacris teach me sensei! ?

EDIT: I suspect the black A12x25 will have crazy prices and yet being sold out for months. That thing is as hotly anticipated as Ampere/Big Navi haha

 

[DrHudacris]

I'm seriously considering doing it for the browns since I have 3 in my builds right now. The 2 that are currently on the GPU will become redundant once I get RDNA2 and put it on water anyway.
@DrHudacris teach me sensei! ?

EDIT: I suspect the black A12x25 will have crazy prices and yet being sold out for months. That thing is as hotly anticipated as Ampere/Big Navi haha

Ahh you first! I can't bring myself to sand them! Plus I'm discouraged with using arctic P12 and still hitting critical temps (with side panels on).

But! I used 3 kaze flex slim with TX240 and my coolant stabilized at 59C degrees with a 330 watt load.

I'm honestly thinking a Cu720 with a Noctua IPPC 3K and slim fans will be super good.

 

[vusun123]

Was kinda disappointed by the fact the host didnt realize that the monoblock wasn't at fault for having higher temps.
The monoblock was also cooling the vrm of the mobo which the alphacool CPU block did not have to do.

Hence the increase in temps. Again, host did not realize it.

I have the exact same setup as Ali was testing in that video (Z390i Strix + 9900K @5Ghz + monoblock) although my CPU is delidded and replaced with copper IHS, and my temp is 6 degrees cooler than his result with the Alphacool block. I was surprised to see his result. I did ask whether or not he delidded his CPU but didn’t get an answer yet

 

[thelaughingman]

Ahh you first! I can't bring myself to sand them! Plus I'm discouraged with using arctic P12 and still hitting critical temps (with side panels on).

But! I used 3 kaze flex slim with TX240 and my coolant stabilized at 59C degrees with a 330 watt load.

I'm honestly thinking a Cu720 with a Noctua IPPC 3K and slim fans will be super good.

from my own experience with the P12, it is a great fan for noise and moderate heat load but it does lack that punch of the top-end of RPM when you really push the power and heat output compare to the A12x25. Your use case falls exactly into that bracket with CPU + dual GPUs on a single rad haha

I briefly looked at the IPPC 3000 but the noise will be unbearable so I'll stick with the A12x25. Bookmarked all your sanding posts so I could refer back later on ?

 

[arthasdklol]

1. LTT IS A HACK

2. Unless they state their Education, THEY DON'T KNOW ANYTHING. Referring to them is like referring to a Wikipedia article without footnotes.

3. Refer to #1

Yeah I don't understand why people take LTT seriously.
It's entertainment, nothing more. He's the Top Gear equivalent of the IT world...

Looking at the PCB sizes and Optimum Tech's video about 3090, I suspect he will consider going Fabio's way of fitting a 120 rad in the spare space due to a shorter PCB

 

[thelaughingman]

Looking at the PCB sizes and Optimum Tech's video about 3090, I suspect he will consider going Fabio's way of fitting a 120 rad in the spare space due to a shorter PCB

TBH I don't think Optimum Tech is going the extreme route, case in point he picked the EK SE Classic rad instead of the TX240 or the more exotic options that many have shared in this thread. His build is more vanilla and there's nothing wrong with that, useful to wider audience who wants to know what's guaranteed to work and safe to build.

 

[fabio]

TBH I don't think Optimum Tech is going the extreme route, case in point he picked the EK SE Classic rad instead of the TX240 or the more exotic options that many have shared in this thread. His build is more vanilla and there's nothing wrong with that, useful to wider audience who wants to know what's guaranteed to work and safe to build.

Exactly. If I wasn't me, I was using exactly the same setup that OptimumTech is using in the latest video. Simple and common parts, no modding. Pretty cheap prices.

 

[thelaughingman]

Exactly. If I wasn't me, I was using exactly the same setup that OptimumTech is using in the latest video. Simple and common parts, no modding. Pretty cheap prices.

yeah you and Huda gave me the crazy thoughts that led me to watercooling and now ultimately to sanding down A12x25s. I love you both SO MUCH ? ? ?

btw what are your thoughts on dual DDC loop in the T1? with the Aquanaut/Freeflow AND the shorter GPU waterblock, it is now possible to cram 2 DDCs in the T1 I think, 1 in each chamber. For sure the redundancy is nice in case 1 pump fails the other pump can keep the loop going. But I wonder what other benefits/drawbacks from such a setup - temps, noise, flow rate, etc.

 

[fabio]

yeah you and Huda gave me the crazy thoughts that led me to watercooling and now ultimately to sanding down A12x25s. I love you both SO MUCH ? ? ?

btw what are your thoughts on dual DDC loop in the T1? with the Aquanaut/Freeflow AND the shorter GPU waterblock, it is now possible to cram 2 DDCs in the T1 I think, 1 in each chamber. For sure the redundancy is nice in case 1 pump fails the other pump can keep the loop going. But I wonder what other benefits/drawbacks from such a setup - temps, noise, flow rate, etc.

ahahahha, I will never sand a Noctua fan! They are egineered almost to perfection, and they use every .mm to reach those performance!

No man, 1 ddc is even too much for this loop! ? If it fails, I will replace it! ?

 

[ten13th]

Exactly. If I wasn't me, I was using exactly the same setup that OptimumTech is using in the latest video. Simple and common parts, no modding. Pretty cheap prices.

The world needs more Fabios, less vanillas.