Ryzen 3950X, 3960X, 3970X Availability

I've had very good luck with EK's translucent fluids.

Two years on one run and everything came out clean as a whistle without any sign of corrosion.

I think after their nickel coating fiasco a few years back, they have been very risk averse with those things, so I expect theirs to be among the better choices.

If still be weary of opaque fluids though.

We're there any warnings on the packaging at all about being "just for show purposes" or something to that effect? I have seen that on mayhem's fluids.

That's the "Mayhem's Aurora", they're crazy metallic silver colors and stuff like that, not just opaque. The pastel lasts longer than their translucent according to them (3 years vs. 9-12 months.).
 
I've had very good luck with EK's translucent fluids.

Two years on one run and everything came out clean as a whistle without any sign of corrosion.

I think after their nickel coating fiasco a few years back, they have been very risk averse with those things, so I expect theirs to be among the better choices.

If still be weary of opaque fluids though.

We're there any warnings on the packaging at all about being "just for show purposes" or something to that effect? I have seen that on mayhem's fluids.

I use clear EK coolant. It does a GREAT job at cooling as well as keeping nasties out. I also have a silver coil in my water tank.
 
I've had very good luck with EK's translucent fluids.

Two years on one run and everything came out clean as a whistle without any sign of corrosion.

I think after their nickel coating fiasco a few years back, they have been very risk averse with those things, so I expect theirs to be among the better choices.

I'd still be weary of any opaque fluids though.

We're there any warnings on the packaging at all about being "just for show purposes" or something to that effect? I have seen that on mayhem's fluids.

https://www.ekwb.com/shop/ek-cryofuel-solid-cloud-white-concentrate-250-ml
 
Making some logos to add to the case...Camera on my phone is potato-quality but you get the picture...(no pun intended)
BOTHLOGOS.jpg
 
USPS just showed up at my door 5 minutes ago with my 4-pack of 1000mm hardlines from Amazon. Radiators are 3 stops away.

Oh...And in breaking news from the "WTF were you stupid motherfuckers thinking?!?!?!" departrment, a very apologetic USPS Postman handed me a box...guesses as to what's inside?
 
  • Like
Reactions: N4CR
like this
USPS just showed up at my door 5 minutes ago with my 4-pack of 1000mm hardlines from Amazon. Radiators are 3 stops away.

Oh...And in breaking news from the "WTF were you stupid motherfuckers thinking?!?!?!" departrment, a very apologetic USPS Postman handed me a box...guesses as to what's inside?

give ya 500 for it :p
 
I use clear EK coolant. It does a GREAT job at cooling as well as keeping nasties out. I also have a silver coil in my water tank.

Both the fluid and the silver coil seems overkill.

I bought a silver coil (and PT Nuke) when I first built my loop, but opted against using them after studying the galvanic series.

After replacing it with my new Heatkiller for the Threadripper, I opened up my EK Supremacy EVO nickel played block to take a closer look (I had previously only peeked in through the inlet and outlet holes).

It's been hard at work in my system since the summer of 2016, so 3.5 years

First year it used EK's old Ekoolant in 2/3's clear, 1/3 blue. Then when I broke in to it to make some changes, I replaced it with EK's newer Cryofuel (hooray for marketing names!), again in 1/3 blue, 2/3 clear.

After I broke into it again last year to replace all of my swivels (I had a couple leak and didn't trust them anymore) I but 100% blue EK Cryofuel in there.

This is what it looked.like today:

IMG_20191222_031649.jpg


It's tough to tell in the picture due to all the reflections and the fact that it was not 100% dry yet, but I saw no gunk when I was l looking at it closely, so it looks like it has done it's job.

Makes me wonder what the people with bad looking blocks did, as I certainly didn't do anything remarkable to keep mine clean.
 
Zarathustra[H]

Its not overkill in the least.

I will skip the text wall response on metal metal interactions, you said you know that stuff. I am assuming most do.

however,

It turns out that additives in coolant that prohibit microbial growth have a rather fast half life and will simply fail to work after so long.

This is called Pharmacokinetics which is a branch of biology dealing with understanding the half-life of anti-bacterial or microbial agents and their applications. This is why a person must take an Antibiotic every 12 hours for 7 days. Its not to outrun the bacteria but it is to ensure that by the time the 7am dose wears off due to half-life the 7pm does can take over again. This allows the body enough time to keep attempting immunological response until the appropriate antibody complex is discovered to clean the rest of the infection out. Antibiotics do not cure the body, they assist. We can apply this same logic to loop inhibitors.

This is why coolants must be changed every so often to remove the biofilms that grew due to the agent halflife. However, adding a silver ion to the water prevents bacteria from growing due to an interaction with the peptidoglycan constituent of the membrane complex of a bacterium. In a water loop silver particles released into the coolant stream and they attach to bacteria breaking down the membrane causing the organism to lyse or break open. Sometimes the ions are returned to the silver through contact by recirculating in the loop.

The silver ion is just another defense against microbial growth in case the bacteria ether grow resistant to the additives or the additives breakdown due to half life. Remember all molecules have a half life, Some are nano seconds, and some like Bismuth are theorized to be ten quadrillion years.

Just for sourcing information, I have degrees in Biology and Chemistry. Doesn't make me a specialist, doesn't make me an authority. Only allows me to understand in more depth how the internet knows so much that isnt so.

I have a gigantic jug of Hydrochloric Acid outside and while I can dissolve Silver and form Silver Chloride in a solution that I can easily get the silver to precipitate back into a solid with another chemical reaction.

However, Silver is NOT soluble in water. So silver will not dissolve in your loop. Silver particles actually release from the silver strip in the water container and those silver micro-particles will release ions into bacteria. The ions are soaked up by bacteria but are eventually released and reacquired by the silver strip to be used infinitely.

I could run my silver strip for 150 years of loops and I might lose 1.0x10^-16 grams of material. Just a fun guess. Its the little $7.00 that keeps on giving and giving and giving and giving until after my death. My son, his Son, his son's son and so on.
 
Last edited:
Zarathustra[H]

Its not overkill in the least.

You must understand galvanic corrosion and the process before ruling something out based on hear say. Silver is like Gold. It is extremely un-reactive and highly stable. It has an extremely low anodic index of only 0.15volts. The same as Gold.

If you look at a common index of strongly to almost non-corrosive metals you see the following:

Ag/Au 0.15v

Ni/Cu .30 to .35v

Al (horribly reactive) is 0.95v because no one uses pure aluminum in manufacturing computer parts

and it goes up to Magnesium etc... @ over 1.75v


By using two highly galvanic metals like Aluminum with something less but still anodic like Nickel as in EK blocks and Aluminum radiators the chances of ionic interactions causing corrosion is significant.

By adding something like EK coolant with anodic inhibitors you prevent the stripping of metal ions from the highly solvent water base that it sits in.


It turns out that additives in coolant that prohibit microbial growth have a rather fast half life and will simply fail to work after so long. Hence leaving you with a sludge in your loop much faster than you should get. However, adding a silver ion to the water prevents bacteria from growing due to an interaction with the peptidoglycan constituent of the membrane complex of a bacterium. In a water loop silver ions are released into solution and they attach to bacteria breaking down the membrane causing the organism to lyse or break open. Sometimes the ions are return to the silver through contact by recirculating in the loop.

The silver ion is just another defense against microbial growth incase the bacteria ether grow resistant to the additives or the additives breakdown due to half life. Remember all molecules have a half life, Some are nano seconds some are trillions of years (in theory).

Just for sourcing information, I have degrees in Biology and Chemistry. Doesn't make me a specialist, doesn't make me an authority. Only allows me to understand in more depth how the internet knows so much that isnt so.


What made me decide against it was the engineering guidelines for harsh wet environments (which I imagine the inside of a loop must be) to keep the differences between the anodic indices of the metals at or below 0.15v.

According to the table I used, silver was -0.15v and copper was -0.35v, which was greater than my target, so I passed on it.

Maybe I was being overly cautious, but at the time I was more concerned about corrosion due to EK's past nickel problems than I was growth. Three years later I have zero growth without any silver, so I guess it worked!

My reading suggests that a simple glycol-distilled water mix is the best coolant. The glycol serves as surfactant, corrosion inhibitor and growth inhibitor all in one, at the right concentration.
 
Last edited:
anyone used photoshop 2020 with your 3950x system? I get lag/floatiness when using it for some reason.. Not all the time its open but consistently here and there.
 
What made me decide against it was the engineering guidelines for harsh wet environments (which I imagine the inside of a loop must be) to keep the differences between the anodic indices of the metals at or below 0.15v.

According to the table I used, silver was -0.15v and copper was -0.35v, which was greater than my target, so I passed on it.

Maybe I was being overly cautious, but at the time I was more concerned about corrosion due to EK's past nickel problems than I was growth. Three years later I have zero growth without any silver, so I guess it worked!

My reading suggests that a simple glycol-distilled water mix is the best coolant. The glycol serves as surfactant, corrosion inhibitor and growth inhibitor all in one, at the right concentration.

I deleted my whole post. You quoted it so it's there if anyone cares Haha. It was just too much of a wall of text and was not needed in this thread. You said you know about Galvanic stuff anyways so I was like meh its not needed in the end.
 
anyone used photoshop 2020 with your 3950x system? I get lag/floatiness when using it for some reason.. Not all the time its open but consistently here and there.

Photoshop doesnt seem to keep up with AMD. Even now that AMD is the leading productivity cpu Adobe still bows to Intel imo.
 
Photoshop doesnt seem to keep up

That's pretty much it for Adobe. They've created applications that are industry standard-bearers and made them accessible, and now they seem to be on the complacent path with record profits.

I do wish they'd focus on taking advantage of CPU (and GPU!) advancements more, as their competition very much has. As it stands, there's a bit of a limit as to what hardware makes Adobe apps faster, despite so many embarrassingly parallel workloads.
 
  • Like
Reactions: N4CR
like this
That's pretty much it for Adobe. They've created applications that are industry standard-bearers and made them accessible, and now they seem to be on the complacent path with record profits.

I do wish they'd focus on taking advantage of CPU (and GPU!) advancements more, as their competition very much has. As it stands, there's a bit of a limit as to what hardware makes Adobe apps faster, despite so many embarrassingly parallel workloads.

Well premier pro has been surpassed by Davinci Resolve 16. I love DR16. So much better and it actually works and doesnt crash even If it has a couple less features.

I love this 60x. Now I'm learning how to use Blender and it looks very powerful for rendering.
 
Last edited:
Well...first this happened
IMAG0731.jpg


I can't sue the USPS itself, but I can sue my postman as an individual since his mis-conduct resulted in a monetary loss for me.

Then I took my mind off that by doing more work on the system. I realized I didn't have a workable mount for the thick radiator on the bottom. The one which came with the case raises the rad too far up so I wouldn't have room for the fan and grill...

So I used the case from hell (TT Core WP200) and removed a section from it...to make a frame for the Radiator
IMAG0740.jpg

Then I riveted it to the case

IMAG0742.jpg


The Radiator is going to breath easily since I removed the bottom slotted section of the case, and the filter still fits fine.

IMAG0744.jpg


I got my PCIe Gen4 Riser Cable (damn that was costly!) and connected the M.2 card and installed it and the 2080 Super (using the Phanteks Riser and vert card mount)

Even with the thick radiator I still have 15-16mm clearance between the bottom Fan Grill and the bottom of the vertically-mounted GPU and M.2 Cards.

I also installed the XSPC Raystorm Pro (problems there...had to make a thread about it asking for assistance cause it has me worried) and called it a day...I need to resolve the CPU block question before I can proceed...GRRRR

IMAG0745.jpg
 
Well...first this happened
View attachment 210275

I can't sue the USPS itself, but I can sue my postman as an individual since his mis-conduct resulted in a monetary loss for me.

Then I took my mind off that by doing more work on the system. I realized I didn't have a workable mount for the thick radiator on the bottom. The one which came with the case raises the rad too far up so I wouldn't have room for the fan and grill...

So I used the case from hell (TT Core WP200) and removed a section from it...to make a frame for the Radiator View attachment 210277
Then I riveted it to the case

View attachment 210278

The Radiator is going to breath easily since I removed the bottom slotted section of the case, and the filter still fits fine.

View attachment 210279

I got my PCIe Gen4 Riser Cable (damn that was costly!) and connected the M.2 card and installed it and the 2080 Super (using the Phanteks Riser and vert card mount)

Even with the thick radiator I still have 15-16mm clearance between the bottom Fan Grill and the bottom of the vertically-mounted GPU and M.2 Cards.

I also installed the XSPC Raystorm Pro (problems there...had to make a thread about it asking for assistance cause it has me worried) and called it a day...I need to resolve the CPU block question before I can proceed...GRRRR

View attachment 210280
Sue him for delivering packages? I can not tell in the pic what is wrong. I see variations, but no real idea what is damaged?
 
Alright.

The routing didn't wind up as good as I'd like due to the criss cross of the input and output ports on the new CPU block, and the fact that the first PCIe slot shifts down one slot length on the Threadripper boards, but it is functional.

IMG_20191222_182120~2.jpg


This was supposed to be a drop in replacement, but instead I'll have to take it apart and re-route it at some point. Maybe when the 3080TI (or whatever Nvidia decides to call it) comes out. Maybe I'll even give hard tubes a try...

Anyway, I'm ready for a smoke test.

I didn't do a pre-build test on this one, as I had nothing but the water block that could cool the Threadripper.

Wish me luck.

Knock on wood, but I guarantee I'll flip the switch and nothing will happen and I'll be troubleshooting 10 minutes from now.
 
Last edited:
Alright.

The routing didn't wind up as good as I'd like due to the cross cross of the input and output ports on the new CPU block, and the fact that the first PCIe slot shifts down one slot length on the Threadripper boards, but it is functional.

View attachment 210318

This was supposed to be a drop in replacement have to take it apart and re-route it at some point. Maybe when the 3080TI (or whatever Nvidia decides to call it). Maybe I'll even give hard tubes a try...

Anyway, I'm ready for a smoke test.

I didn't do a pre-build test on this one, as I had nothing but the water block that could cool the Threadripper.

Wish me luck.

Knock on wood, but I guarantee I'll flip the switch and nothing will happen and I'll be troubleshooting 10 minutes from now.

#1 Damn, no PS2 keyboard port on these here newfangled motherboards :p

(It's been a while since my last upgrade :p )
 
#1 Damn, no PS2 keyboard port on these here newfangled motherboards :p

(It's been a while since my last upgrade :p )

#2 Eww. Mouse driven GUI BIOS.

Now I'm going to have to find a mouse. Not just any mouse, but a USB mouse...

I guess this one just won't do anymore...

IMG_20191222_185342.jpg


If anything this upgrade is teaching me that it is time to update the spare accessories in the basement I use for builds and setups :p

But we DO have a bios, which is a success! It scared the shit out of me though, because it took for bloody ever for the Post screen to show.

Hopefully this is a "initial power on only" thing
 
Last edited:
Sue him for delivering packages? I can not tell in the pic what is wrong. I see variations, but no real idea what is damaged?

You would have to read my earlier post to fully understand...The package was supposed to have been delivered 9 December. It was marked as "Delivered. In Mail Box" on 9 December. It was never actually delivered until yesterday afternoon...11 days after he falsified the record to say it was delivered.

It's against USPS regulations to sign a package as delivered if it wasn't. He had 2 options...bring it to my house or leave a pink slip in my mailbox stating it was being held at the main facility. He did neither. USPS considers falsifying the delivery notice as misconduct. This opens him up to being sued

For 11 days he told no one at all. (I assume he didn't say anything to anyone until after I opened a mail fraud case with the Postal Inspector's Office last week)

Best Buy sold that processor at MSRP. Because it was never delivered, I had to go elsewhere to find a replacement, and pay a $200 premium over MSRP to get it from a place which will not allow me to return it, they'll only replace it (with a re-stocking fee.) I also took leave on the 9th to be there when the CPU was delivered (and it wasn't)


So...I'm out $300 for a lost day of work
I'm out a leave day
I'm out $200 over MSRP I had to pay to get a replacement

His excuse was, "I didn't want to leave it at your door where someone could take it."

Had he attempted to do EXACTLY THAT, he'd have been greeted with a large note stating, "I'm Home USPS Guy! Just knock!. Thanks!!!", thereby avoiding ALL THIS crap...

Besides my monetary damages, Best Buy had to file a claim with their insurance to cover the $800+ "lost" CPU (because they refunded me last week for something that wasn't delivered)
 
  • Like
Reactions: N4CR
like this
You would have to read my earlier post to fully understand...The package was supposed to have been delivered 9 December. It was marked as "Delivered. In Mail Box" on 9 December. It was never actually delivered until yesterday afternoon...11 days after he falsified the record to say it was delivered.

It's against USPS regulations to sign a package as delivered if it wasn't. He had 2 options...bring it to my house or leave a pink slip in my mailbox stating it was being held at the main facility. He did neither. USPS considers falsifying the delivery notice as misconduct. This opens him up to being sued

For 11 days he told no one at all. (I assume he didn't say anything to anyone until after I opened a mail fraud case with the Postal Inspector's Office last week)

Best Buy sold that processor at MSRP. Because it was never delivered, I had to go elsewhere to find a replacement, and pay a $200 premium over MSRP to get it from a place which will not allow me to return it, they'll only replace it (with a re-stocking fee.) I also took leave on the 9th to be there when the CPU was delivered (and it wasn't)


So...I'm out $300 for a lost day of work
I'm out a leave day
I'm out $200 over MSRP I had to pay to get a replacement

His excuse was, "I didn't want to leave it at your door where someone could take it."

Had he attempted to do EXACTLY THAT, he'd have been greeted with a large note stating, "I'm Home USPS Guy! Just knock!. Thanks!!!", thereby avoiding ALL THIS crap...

Besides my monetary damages, Best Buy had to file a claim with their insurance to cover the $800+ "lost" CPU (because they refunded me last week for something that wasn't delivered)
Ahhh! That makes sense. Ouch. That is wrong. Yeah, he needs to be talked to!
 
You would have to read my earlier post to fully understand...The package was supposed to have been delivered 9 December. It was marked as "Delivered. In Mail Box" on 9 December. It was never actually delivered until yesterday afternoon...11 days after he falsified the record to say it was delivered.

It's against USPS regulations to sign a package as delivered if it wasn't. He had 2 options...bring it to my house or leave a pink slip in my mailbox stating it was being held at the main facility. He did neither. USPS considers falsifying the delivery notice as misconduct. This opens him up to being sued

For 11 days he told no one at all. (I assume he didn't say anything to anyone until after I opened a mail fraud case with the Postal Inspector's Office last week)

Best Buy sold that processor at MSRP. Because it was never delivered, I had to go elsewhere to find a replacement, and pay a $200 premium over MSRP to get it from a place which will not allow me to return it, they'll only replace it (with a re-stocking fee.) I also took leave on the 9th to be there when the CPU was delivered (and it wasn't)


So...I'm out $300 for a lost day of work
I'm out a leave day
I'm out $200 over MSRP I had to pay to get a replacement

His excuse was, "I didn't want to leave it at your door where someone could take it."

Had he attempted to do EXACTLY THAT, he'd have been greeted with a large note stating, "I'm Home USPS Guy! Just knock!. Thanks!!!", thereby avoiding ALL THIS crap...

Besides my monetary damages, Best Buy had to file a claim with their insurance to cover the $800+ "lost" CPU (because they refunded me last week for something that wasn't delivered)

Sounds really fishy.

Almost as if he was thinking of stealing it, but had an change of heart.

Either that or he is brand new, and doesn't have a clue.
 
  • Like
Reactions: N4CR
like this
Sounds really fishy.

Almost as if he was thinking of stealing it, but had an change of heart.

Either that or he is brand new, and doesn't have a clue.

The really fishy thing about it was that he delivered my 4-pack of hardlines and left...then I saw him on my door bell camera putting a package outside my door so I opened it before he could leave and that was the CPU...that's when he was all apologetic...

I tried to update my USPS case online to let them know about it but it says I have to call them because the investigation is open...that's going to be a VERY fun call tomorrow.
 
The really fishy thing about it was that he delivered my 4-pack of hardlines and left...then I saw him on my door bell camera putting a package outside my door so I opened it before he could leave and that was the CPU...that's when he was all apologetic...

I tried to update my USPS case online to let them know about it but it says I have to call them because the investigation is open...that's going to be a VERY fun call tomorrow.

I wish you the best of luck. I'm glad you are pursuing it. Accountability is important!
 
Alright.

The routing didn't wind up as good as I'd like due to the criss cross of the input and output ports on the new CPU block, and the fact that the first PCIe slot shifts down one slot length on the Threadripper boards, but it is functional.

View attachment 210318

This was supposed to be a drop in replacement, but instead I'll have to take it apart and re-route it at some point. Maybe when the 3080TI (or whatever Nvidia decides to call it) comes out. Maybe I'll even give hard tubes a try...

Anyway, I'm ready for a smoke test.

I didn't do a pre-build test on this one, as I had nothing but the water block that could cool the Threadripper.

Wish me luck.

Knock on wood, but I guarantee I'll flip the switch and nothing will happen and I'll be troubleshooting 10 minutes from now.


If you decide to give hard tubes a try later, I do recommend picking up an extra 4-pack of the 500mm hard lines. You can then practice on 1 of the tubes to get the hang of heating/bending them and have 3 left over in case you need to re-do something (like your current situation, for example)

I have 2 of the black Noctua Chromax 140s on a shelf, along with several of their Beige & Red beauties. I also have 24 Corsair Red & Blue fans leftover from the case from hell build. (I replaced all 24 with Noctuas (first time I ever made a $700 order for just fans, lol)
 
Temporary setback.

Plan was to transfer my images to my new NVMe drive (installed and buried under GPU) by booting from a Linux live image.

I can't find a Linux live image with a new enough kernel to boot on the Threadripper.

Trying to install Mint from live stick to another USB stick on my laptop, boot up from the USB stick, install the latest kernels and then try booting it on the Threadripper.

I knew this was going too easily there for a bit :p

Edit:

Never mind. It was the USB controller that was the culprit. Booting off of a USB2 port worked just fine.
 
Last edited:
Temporary setback.

Plan was to transfer my images to my new NVMe drive (installed and buried under GPU) by booting from a Linux live image.

I can't find a Linux live image with a new enough kernel to boot on the Threadripper.

Trying to install Mint from live stick to another USB stick on my laptop, boot up from the USB stick, install the latest kernels and then try booting it on the Threadripper.

I knew this was going too easily there for a bit :p

Edit:

Never mind. It was the USB controller that was the culprit. Booting off of a USB2 port worked just fine.

Glad to hear it worked out!
 
Alright.

The routing didn't wind up as good as I'd like due to the criss cross of the input and output ports on the new CPU block, and the fact that the first PCIe slot shifts down one slot length on the Threadripper boards, but it is functional.

View attachment 210318

This was supposed to be a drop in replacement, but instead I'll have to take it apart and re-route it at some point. Maybe when the 3080TI (or whatever Nvidia decides to call it) comes out. Maybe I'll even give hard tubes a try...

Anyway, I'm ready for a smoke test.

I didn't do a pre-build test on this one, as I had nothing but the water block that could cool the Threadripper.

Wish me luck.

Knock on wood, but I guarantee I'll flip the switch and nothing will happen and I'll be troubleshooting 10 minutes from now.

LOL you told me my little silver coil was overkill but have you seen that front radiator lmao hahaha

Just picking at you man

Anyways that long delay in post was from ram training. It will be locked in the eprom once its figured out the training.
 
LOL you told me my little silver coil was overkill but have you seen that front radiator lmao hahaha

Just picking at you man

Anyways that long delay in post was from ram training. It will be locked in the eprom once its figured out the training.

I don't like fan noise.

More radiator capacity = less fan speed :p
 
Well, I thought I'd be able to get my existing Mint 19.2 install with the Ubuntu LTS Hardware Enablement 5.3 kernel installed to boot, but it is flat out refusing.

Funny, since the 5.9 kernel boots just fine from the USB stick.

May have to do a clean 19.3 install.

This is turning out to be a bigger project than I had anticipated. Especially since the Asus XG438Q just arrived...
 
Back
Top