if it performs well enough, i don't think you'll lose money (although obviously you wont know how well it'll sell til you try, unfortunately) and even if its only on a par with a maze 3, thats still pretty good from something thats nice and compact and looks relatively cheap to make (since its lathed, rather than all machined)

yep, i'm the same bloke, thought i'd use the same name here so people know who i am (same name on OcAU too :) )
unfortunately i wont have time to test it at the mo cos i've got university exams going on next month, but if you still want it tested in about a months time, i'd be happy to (though i'd expect you would have tested it yourself by then, and maybe have made some more. i'm sure i'd be interested in getting hold of one of the final blocks :) )

one other thing i thought of, are you going to put holes in the top for athlon-64 mountings as well, or does that complicate things? just thought it'd add some future-proofing to the design

i'd say its more likely that its just a coincidence with the DD design, not much you could do about it though even if they had nicked it, since i guess your design isn't copy-righted or anything :( just the way it goes really, i wouldn't worry about it, its not as though they're identical (yours might even work better :D )

No, they didn't steal the design. They have been working on it for a lot longer than this thread has existed. The one in the picture is months old and just a prototype. The retail version I hear is modified quiet a bit from that.

Either way, I havn't been more angry than this in a long time.

I can't believe that I put so much effort into this design, and its just been ripped by DD.

Depresssing is not the word for it.

FWIW...

I think that this Maze4 is a natural progression on the Maze3: I wouldn't beat myself up over it. We've been waiting for something new from DD for a very long time.

i really don't think they copied it, companies work on things for a long time before bringing out new stuff, its just a coincidence that it looks similar to yours, i don't think its anything you should be too worried about really

Well, no-ones gonna buy my block when DD have one almost exactly the same are they :(

Nah, you're probably right, they didn't steal it, but its the worst possible thing that could have happened, a company, probably the most respected in the whole watercooling world, has a block almost identical to mine, and there's nothing I can do to compete.

depends how much yours costs i guess, i seem to remember you estimated a price of around 25-30 quid? still looking at about the same price? cos the maze 4 is probably going to be 40-45, like most commercial blocks. people will buy yours if you stick to that sort of price range i'm sure, so i don't think its any reason to give up on the block :) otherwise i guess you could try and think of ways of moddifying the design slightly so that it looks less similar to the DD one

I don't know about that... it was Swiftech after all that hired Bill... and D-TEK that purchased the rights to the White Water... seems to me that gives both of those two have a head start on Danger Den... ;)

Danger Den has been around longer and is more well know to the newbs and their stuff is less expensive. I try to get people to buy from DTek almost everyday on the forums I browse and they usually end up buying a DD block because their friend recommended it and it is cheaper. :rolleyes:

I don't think it's fair to say they're the most respected water vendor just because all the noobs buy there. I do know what you mean though... whenever a "what's the most l337 block?" thread pops up at Overclockers the Maze 3 always gets high praise... sometimes it even comes "within a degree or two" of the White Water. ;)

Don't completely disagree with any of that, except DD was one of the first and still one of the best, and continues to innovate. They deserve a level of respect. At least IMO. I personal think Dtek is the best though. Best blocks period! They got number 1, 2, and 3 IMO.

Here is the production version.

They've used a mill to make that block (and the NB/GPU block) so you could cut the cost(and time?) by using a lathe could'nt you...

Don't give up on your block, they are'nt the same really, just simular. Performance is based on surface area, baseplate thickness, ect. Not looks...

Comparing their block to mine...

1. They have 4 grooves, I have 6/7 if you include the middle hole
2. Their grooves are wider than mine
3. Their block is oval, not circular
4. Their grooves look VERY shallow, which is very odd, I worked out a depth of 5mm would probably be best in this sort of design.
5. It looks like the base on the block is fairly thick, I still think they could shoot for something like 1.5-2mm, I'm going for 0.8-1.2 on mine.
7. Their grooves don't continue to under the inlet and outlet... thats quite interesting. I don't know if thats a good or bad thing, I'll thinkabout it.

Well, if their Maze4 beats their Maze3, then mine should beat both quite handily, and hopefully not cost an awfull lot more to make, there might be life left in my design yet.

And I say respected not on the ground that n00bs buy them, or that they're particularly high performance, but that they have a very good reputation for reliabilty, and there are very few things to fault in their blocks. They make cheap, decent performance blocks, and as a company, they seem freindly and customer orientated.

For example, they did have a slight issue with the clear tops on the Maze3 when they first came out. They fixed the problem, but also replaced many blocks, free of charge, that had had problems. To me, thats the mark of a respectable company.

I know what you mean though, there are many ways to measure respectability, but to me, they do very good work.

( I'm less pissed off now, after some sleep, was a long week at work :p )

It's purely numbers with DD IMO. SO MANY people brought them because it was them or Swiftech (I did), availability plays a big part as well. They are 'respected by numbers' not by performance ;) ...

Well, I've tested it, and results were impressive, beating a Maze3 by 3-4 degrees.

Of course, and as I'm sure someone will leap in a point out, my testing cannot be relied upon, too many discrepancies etc. etc. :p

Either way, its looking good, and I'm going to find some way to get it tested properly after I make a few modifiactions.

The only problem is, its a HARSH flow killer, in its current form at least. The middle plate, with the slits, either needs to be eliminated, or modified. I know that flow isn't that important in some cases, and as Cathar rightly points out, worrying about flow restriction may be holding back waterblock design.

However, I use my watercooling on all the other hot parts in my system, and I need to save SOME flow for the other 4 blocks I use, so I'm going to sort it out somehow.

In either case, I feel that the basic concept of the block has been proved, and proved to work well, but it needs refining slightly in order to make it possible to sell, and more suitable for the average user.

I'll keep you posted, but for now, its back to the drawing board.

What if you just made a slit where the barbs are except in the base itself instead of the middle plate. Would drop costs a bit if it worked.

the slit exists for one reason (on your design):
To spread the flow of water to all the channels as evenly as possible.
Its not there for impingment, is it? Because if it is, its not over the core. I'd like to see how it works with NO middle plate at all.

@Jaydee

That was my original plan, and the problem was that it has to be fairly narrow, so as not to reduce surface are too much, and then you have the problem of spraying water into a 2mm-3mm wide slit, and then having that slit suppling water to 4-6 grooves ( well, 2-3 grooves, but the water flows in both directions around the grooves ) in either case, the cross section of the "supply line" is about 1x5, and the croos section, in total, of the grooves the water the goes into is about 6 times that. This would mean that the water would tend to use the outside channels a lot more, where the inlet is directly over the channels, and is not relying on the slit to distibute the water.

I could be wrong on this, please feel free to correct me.


@Althornin

Yeah, that why the slits are there. However, some of the area directly under the slit is actually directly over the core. Anyway, with no middle plate at all, the innermost 2-3 channels would not be used. This is due to the fact that there is a limit to how close together the barbs can be, due to you having to be able to fit pipe over them. In either case, there has to be some method of splitting the water flow, or some areas of the block are not going tobe used.


I was considering something like this, inspired cheakily by the Maze4 :p

http://www.gta45.pwp.blueyonder.co.uk/newbase1.jpg

The orange areas are the upright sections, at height 0, balck sections are at height -5mm, grey section is for the o-ring, say -2mm deep.

The larger black circlesshould distribute the flow, and be very cheap to make. This also eliminates the need for the middle plate alltogether, which should keep costs down, and make for a claner looking block.

Opinions?

Sorry not got time to read it all, and someone might have suggested this already, but from what ive seen youll be better making the holes in the centre plate triangular rather than oblongs, thinner the closer to the edges.

Only saw this from the ocuk reports soz m8...

GL

]JR[

Ala,

http://www.phjrw.34sp.com/gta.jpg

Obviously dont make the fat end wider than the smallest circle, or if you really wanted to you could make it logaritmic curved too ;)

]JR[

if you think that the outer channels are getting more water than the inner channels, i agree with the idea given ]JR[, but that would increase the milling coast i think, but by a little.

now another thing, you know, your block reminds me of my maths book :cry:

Lol, as if what's inside isn't bad enough - at least my first year calculus book had a nice sunset or something so when you were studying you could just close and relax. Talk about incentive to keep the book open :D

Nah the point being the closer to the core you are the higher you want the flow to be, so by triangulating the holes you make a lower impedance route for the water in the middle as opposed to the outside.

And hey presto you should have high velocity in the centre, decreasing as you go o u t w a r d s in rings...

edit for some reason it kept ***'ing my ******ds.

]JR[

it is the same thing i thought of and the one which made me agree with you, so why the "Nah"??? :shrug:

:( sorry misread

All good, my bad :)

]JR[

no problem man :) the world is still turning around :D

The only issue with the triangular slots, which I considered and discussed with the machinist, is the cost.

Yes, its the way to go, if cost is not an issue, but it would require them to make a punch, custom made for doing this, which would set me back several hundred before the first block was even made.

This is not a cost I can support, so I'm not doing it. I rejected that idea early on, but well done, very well done for spotting it.

I'm still in favour of getting rid of the middle plate altogether, as it would save me £5+ per block, and as it looks like I will be competing with the Maze4, cost is a definate issue, I need to undercut them by a reasonable amount.

What do you think of the design above? With the extra circles in the base, underneth the inlet and outlet?