[tijgert]

So I got myself a sweet Asus Strix 980Ti with a pretty shiny EKWB block. Having had several EKWB blocks in the last few years I thought I knew what kind of quality to expect, but not this time.

Upon inspecting the craftsmanship I was unpleasantly surprised by extreme machine marks on the contact area's of the water block.
The backplate wasn't exactly smooth either but had some mirror quality surfaces together with rough ridges. It is clear to me that for some reason the Asus Strix 980Ti products, or at least the ones I received, have not had any polishing or buffing done to them and are just like they came from the machine, with a coarse setting. This to the detriment of the cooling performance.
Could this be a rush job or a product given less attention to for being in a niche...

An average air cooler would never get away with such a coarse surface (looking at a Scythe Mugen 3's surface while typing this that reflects my face real nice). It's also unfixable as the nickel plating prevents me from buffing out those machine marks without peeling off the nickel.

Can anyone comment of whether I am just too demanding or should return this set and request a proper finished product?



p.s.
I do not accept that due to the great performance of a block like this reduced performance is negligible and/or should be acceptable. I pay top Dollar (Euro) for a product like this and do not accept short cuts or anything less than the best product possible for my money.

 

[Prophet4NO1]

RMA, seems like a simple issue to deal with.

 

[NKrader]

did you even message them about their mistake?

 

[tijgert]

RMA makes sense, I'm just wondering if I'm too demanding or not. They haven't responded to my mails so I had the reseller contact them. Still no reply.

 

[MrTOOSHORT]

Too demanding?, no.

RMA that piece of crap and get a properly machined block. But really, what a waste of your time.

 

[Jakusonfire]

All blocks have machine marks on them. This isn't jewelry.

Horrendous quality is nothing but hyperbole.

 

[toolmaker03]

return it and get a bitspower block for that video card

 

[tijgert]

I know it's not jewelry, but mark's like these would be an absolute fail on any high end cooler I think.

Asus Strix 980Ti water blocks are few and far between. If I were to decide to deal with the issue myself, would peeling the nickel off be a solution?
Could I Dremel that surface and simple have bare copper after I abrade the plating?

 

[tiborrr12]

Hi,

this is a pattern of a fi 43mm diamond cutting head to make the surface absolutely flat and smooth. It just reflects differently than the rest of the water block which is machined using regular widia bits.

All coolers are made like this except for CPU water blocks which are also further hand polished (hence these marks are removed).

Regards,
Niko

 

[tijgert]

Hi and thanks for taking time to reply personally (I knew there was a rep around).
A GPU uses a lot more power so I don't see why a CPU block surface does get the polish treatment and the GPU block surface doesn't.
I base my disappointment on my far better previous experience with EK blocks and the absolute shine that (indeed) my CPU air cooler has, I can see my face in that.

I'm looking now for alternate blocks, Bitspower being a contender but like I said Asus Strix 980Ti blocks are rather rare, and as soon as I see a more shiney surface on any block, they will get my money. (The fact that they supply Fujipoly speaks for them quite well as well).

We wouldn't be having this discussion if you made plane copper blocks so I could buff it out myself...

 

[tiborrr12]

Hello tijgert,

sorry for dissapointing you this time around

But the fact is this the best possible surface treatment we can offer, but diamond cutters don't make the prettiest cuts. This 'ugly' surface treatment is actually of the reasons why the block performs so well.

The surface is indeed ultra-flat, it cannot get any better. Visually it may not be the best looking, I agree, but most importantly it's dead flat and way flatter than it was ever before in the history of our company.

Hand polishing is out of the question here because of all the different pedestals around the GPU contact surface. Hand polishing or even machine polishing may make the surface uneven.

Fujipoly thermal pads are just like any other brand thermal pads made from the same compound. Every company offers everything from ultra-soft (and somewhat bad thermal performance) to best performance pads (extra stiff). There's always a compromise between compressability and thermal performance. We believe we offer best trade-off in both.

Bulk thermal performance figure has - contrary to popular belief - very little to do with actual thermal performance of the TIM material.

Again, sorry if you are feeling let down at the moment, I hope this post explains the situation from our perspective.

Thanks,
Niko

 

[mattliston]

OP, you need to realize something that has a perfect mirror finish is actually not perfectly flat. it will have the slightest of curved finishes.

seeing machine marks like that could be as minor as 1/1000 of a milimeter, and while able to see a reflection difference, ZERO TEMPERATURE difference may exist.

I highly suggest TESTING it.

RMA'ing something purely on SPECULATION on a cooling device is foolish and wastes time and money on BOTH ends of the transaction.

You will still have the full ability to RMA it if it tests out of spec.

I see this pattern a lot while dealing with machined cylinder heads and blocks. sometimes the inside looks HORRIBLY curved or warped, yet even on a certified straight edge that is accurate to 1/1000 of an inch, still tests out flat.

Perception is occasionally something you need to ignore, and dig right into testing.

Just my opinion and advice

 

[tijgert]

In all fairness, they're valid points...
I knew shiny doesn't mean flat and that flat doesn't mean shiny, I just had a very hard time believing that THIS un-shiny would still mean smooth enough to give top cooling performance.

I guess I'll try this trade-off of smootheness for flatness and see how that goes.
Appreciate the explanation though, puts my heart a little at ease for now.

 

[hemirunner426]

The machining marks are not a big deal. As the rep said the GPU pad is perfectly flat it just hasn't been polished. Polish is not required for a flat surface void of pits and defects.

I have the same block for the 980 strix and the pad had the same machining marks. If you fingernail doesn't catch any of the marks, your block is good to go.

Temps are 0C delta between GPU and water temps at idle. During a session of battlefront there is about a 5C delta between the two.
The most dramatic temp difference between your block and the stock cooler is VRM temps. Stock cooler put the VRMs at close to 90C at load. With the EK block, they see a max of 42C at load.

 

[Kinaesthetic]

Quote:

Originally Posted by tijgert

In all fairness, they're valid points...
I knew shiny doesn't mean flat and that flat doesn't mean shiny, I just had a very hard time believing that THIS un-shiny would still mean smooth enough to give top cooling performance.

I guess I'll try this trade-off of smootheness for flatness and see how that goes.
Appreciate the explanation though, puts my heart a little at ease for now.

Ironically, if anything it should actually improve the performance of the block since any micro-variations in the surface of the block would actually be providing MORE surface area for heat transfer.

Also, this seems like an incredibly silly thing to complain about considering you are going to see it all of about 10 minutes while putting the block on your card.

 

[Alastair]

Quote:

Originally Posted by tijgert

In all fairness, they're valid points...
I knew shiny doesn't mean flat and that flat doesn't mean shiny, I just had a very hard time believing that THIS un-shiny would still mean smooth enough to give top cooling performance.

I guess I'll try this trade-off of smootheness for flatness and see how that goes.
Appreciate the explanation though, puts my heart a little at ease for now.

I think the rep has given pretty valid reasons for why you are seeing what you are seeing. And if you aren't happy I personally think go buy yourself another block and sell the EK one. But don't go trying to RMA the EK one when it is in perfectly working order.

And FWI while it may not be a 980ti strixx block. My Fury X blocks from EK are copper and have the machine marks you speak of. But the performance is stellar. They reach 40C on each card with 1200MHz with +125mv with 680mm of radiator. Also taking into account that I also have a 5GHz FX-8370 in the same loop as well. I honestly think you are being a bit nitpicky here.

 

[Artikbot]

Can you feel the imperfections with a credit card, edge of a paper sheet, or similar?

If you can't it's just your eyes playing you a nice trick and not something you should worry about.

 

[tijgert]

When I said "In all fairness, they're valid points..." I was referring to the comments of both the rep and 'mattliston', NOT to any point I was making myself.
I am giving them the benefit of the doubt / borderline agreeing with them.
Basically, if the rep doesn't freak out about the pic than surely it must be normal.

I think this case has been put to rest and thanks to all for chiming in.

 

[Jakusonfire]

Quote:

Originally Posted by Artikbot

Can you feel the imperfections with a credit card, edge of a paper sheet, or similar?

If you can't it's just your eyes playing you a nice trick and not something you should worry about.

It's purely optical. You can run your finger or fingernail over it and feel nothing. Hell, you can run a razor blade or scalpal over it and it will not catch anything. It's only really visible if you angle the surface just right.

 

[tijgert]

Well by Jove! It really does not catch on anything. It sort of scrapes though, most defo not the sliding across a mirror finish, but not as poor as it seems to the naked eye.
The backplate does have edges it catches on, but that's for the thermal pads to overcome and not nearly as critical.