Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.Daz 3D is part of
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
You're right, but it helps to have a scientifically-based shader anyway =) So that you can get the metal's Fresnel curve (that gives a "pure" metal its colours that vary ever so slightly with viewing angle instead of being all uniformly tinted) and an easy way to get the oxide film (more colours - those blues etc).
Victor and Co have those beautiful vids up:
https://vimeo.com/groups/3delight/videos/147743724
And here's a test of mine... much less cool-looking, but the point is that for the end-user who writes shaders, it is extremely easy to implement this functionality with minimum code:
Morning, well I have a few things I need to do as soon as I can make some noise outside.
As y'all can guess, I'm not one to worry much about 'legacy' color schemes, tho I do worry about how things keep there cool. And that dose include dust build up, lol.
I have a few things I need to attend to with the new GTX960 card where it is, mostly to do with airflow. I think I'm just going to remove that red bit of plastic on the HDD rack and leave the black one. The right most guide vane guides air threw the hard drives, the red one is now restricting the GPU fan and needs to move. Also, with the back blocked below the SATA control card by the GTX960, it has a bit of a doldrums that needs to be addressed by cutting some vents in the SATA control bracket (after disentangling the bracket from the card). So that is the plan for this morning. Clean, move a baffle, and cut some holes.
Well, it looks like the universe is not going to cooperate for blowing the dust out of the computer, as usual. Tho I still need to get the air flowing threw the computer.
Yes, that is rain clouds rolling in.
Kettu, silly question, is that even in Daz Studio, or stand alone 3DL? How long did it take on what CPU for each frame to render that vid? It looks incredible.
(Sarcastic prod at Daz) I agree to an extent with PBR surface settings, as long as the shader implementation is 'Reasonable' for the computer it is intended to run on. I feel the pear pressure for true surface simulation has gon to far when things like the AoA subsurface scatter on HD figures takes more compute power to do then it takes to simulate a hurricane path weather simulation, especially when the program the shader was configured for (Daz Studio) dose not even run on a supercomputer. Yes, knowing how metals and other surfaces react to light is good, tho when the intended customer is ordinary people with a PC, you must be reasonable with what your shader is asking of the computer, lol. If your shader is for entities that can afford thousand core computers then that is fine, tho I don't think that is a majority of Daz3d's costumer base. It took me over a year to scrape up for a GTX960 just to be able to work with simple Iray objects, I will probably never be able to afford a CPU/motherboard combo capable of working with HD figures (If they ever make one that is not for a server or larger computer). Yes, Kaby Lake and Zen are looking good, so long as you do not need Ten Times more CPU power then the top of the line 8-core 16-thread i7 has for spot-rendering scenes with 3delight. AoA shaders on HD figures is just not usable on a PC in it's current implementation in Daz Studio.
A tad bit of off the cuf chat about this morning. I did look at that former photo of the cards in the computer for over a week before doing any thing, as cutting holes in some things can not be undone, and it can break stuff if done wrong. The first thing I did was to mark the PCI opening of the case onto the card's back-plate before taking the card out of the computer. I don't care about marks there as the only thing that looks at the back of my computers is the wall, lol. Something other then a sharpie may be proffered if the metal will be more visible, and don't just dive into it without a backup plan if the item is completely destroyed by a mistake.
I decided to only cut out where the card is, just to force the air to cool something before escaping out the back of the computer. After marking up the metal a bit more, I put the screws in a zip-lock bag and the PCB in a static bag to keep them safe while I waited for the rain to pause. Then I went to look at the HDD rack and decided to just fold the one "turning vane" over to get it out of the way of the GPU fan. I can remove it later if it needs to be removed.
After a fit test, I decided to leave that as is. I then took the SATA card back plate outside to work on it away from the computers, just to keep the metal dust away from delicate electronic stuff. These things have a ten-thousands of an inch tolerance, bending it is very bad. Let the grinding wheel cut the metal, not your muscles.
This is where I should suggest a good vid that covers a good way to get burs and metal splinters off of a cut metal edge. It need not be as precise as what AgentJayZ did with the jat blades, tho the process is quite similar if not exactly the same to this.
http://www.youtube.com/watch?v=iLSrgZIDAQk
That followed by vacuuming off the part, and your hands, cloths, and any thing else that will be going back inside with you to the computer. I further used a paper towel to wipe off the bracket just to be sure there was no metal dust that could get under a BGA chip in the computer.
Then, after I was sure that there was no metal bits that could fall off and short something out, I put the card back together.
I think That is going to breath a lot better then the graphics card now
Well, I'm still looking at heat sinks.
I made a close proxy in hex to get an idea of what is going on in that GTX960, and I'm just not sure what to do with it yet. I tried a few different baffles around the thing to coax more cooling threw the thing, tho I'm not all that surprised by the lack of results.
The box to force air threw the card from the bottom of the case did nothing at all for temps. And the vain on the front of the card to keep the intake fan from having a sumo match with the GPU fan actually made things worse.
Now I'm quite tempted to cut some holes in the plastic cover on the card to let more air exit near the back of the card, however that would only be blasting hot air down at my hard drives in the best of circumstances. Because it is a Positive pressure computer, it is more likely cool air will force it's way out there reducing the air flow threw the heat sink akin to a Compressor stall in a jet, not good.
There is a small amount of gap between the heat sink and card in some spots that I could open up, however it is not much to be honest.
So I guess I'm going to think about this a bit more, and hope the thing dose not melt the motherboard in the mean time. Because THG is not that far off with the temps of this PCB layout.
http://www.thg.ru/graphic/obzor_nvidia_geforce_gtx_960_part_2/print.html
I have no idea where the English version went, sorry y'all.
It looks like you're pre-heating the air by passing it around the HDD before it gets to the GPU!!!
Try ducting in cool air directly to the GPU from outside the case, or increase the exhaust from the case in general, or both.
yea, it's kind of difficult to see what is going on in photos, as much of it is 3d layers. And by the way that is why the other two HDD racks are sitting on the floor outside the PC, hence all the wires going out the back of the computer, to test just that. Originally I had that wide open so the card could get air from anywhere.
The GPU is getting three PCI slots of fresh air directly from the intake fan, and you can probably guess why I tried that scoop idea after looking at that.
Some things that should have made temps better, just made them worse. As for the pre heating bit, yes and no. Its a 120 watt TDP graphics card getting some air from a 7 watt TDP HDD. And that 7 watt HDD has a thermocouple on it controlling the intake fan, so the hotter it gets the more air it gets, so over all air temp rise at the graphics card is not that much, even with that rack sitting on the floor outside the computer (yes I tried that).
Ya, now, I've gotten some grief on a forum else where over going all out on this IBM 8088 server case to only turn around and "Drop the drives on the bottom of the case". There was a method to that insanity, mechanical isolation with the foam feet on the rack, and eas of pulling out the drives while working on stuff. In any case, completely removing the hard drive rack did nothing at all for temps, so the culprit is not getting hotter air, the card just is not cooling it's self as much as I would like regardless of how much fresh air I give it.
So that is the temps, with things set up like this...
BRB, I need to take a pic of it in the computer, and I need to grab a heater from out back (It got a tad chilly last night).
Umm...with those temps you posted earlier, I thought you were talking about a heater!
lol. Na, the heater is for the pipes just encase they freeze tonight. Nothing like wanting to make coffee and not having running water. The heat tape is plugged in (now), tho I'm not sure yet if it is working. The heater was buried in the back of the shed. In any case, this is what the computer looks like.
Yea, I've yet to make a SSD rack for that pile of drives up front.
At least the GTX960 does not sound like a J79, lol.
http://www.youtube.com/watch?v=YXo5awLzkKU
I had two 6800GT cards at one time, and the beast did sound like that in 3dmark. It kind of made working with audio difficult even tho it sounded incredible.
OK, here is the heatsink and shroud for the EVGA GTX960SC, five #0 Philips head screws, and a slight flex of the bracket to get the shroud to clear the back bracket. Not that difficult for getting in there to get persistent dust out. I 'think' Gamers Nexus had done a complete teardown of a similar ITX card with the same cooler, I'm fairly sure anyway.
(EDIT) Looks like my efforts were in vain, as the card was set by default to "GPU boost" up to 80c or 100% TDP. So no mater what I did around the card to feed it more air, it would just defy me and stay hot as all hell nailed up against the max V-limit.
Some minor progress, tho I have some doubts about the background load of the software (I'd rather it be in BIOS settings, rather then sucking up CPU cycles).
I set the power limit to 75% (90 watts), with a temp target of 60c. Gave it a fan curve, and that's it so far. IF, emphases on the 'if', the GTX1070 can be "leveled from orbit" like this, it may just be usable after all, without sacrificing the 8GB of DDR5 for Iray. Time to fuss with that fan curve a bit more.
Looks like the fan speed thing only goes when the program is open sucking up CPU cycles, hmmmm.
Yep, close the program and the fan stops completely, I was hoping for something a tad more effective then that.
Drop the TDP and temp limit, and get a faster clock out of it. Fascinating
Nope, it's reading right man. Then your not reading it right! lol.
I have a dream, that some day there will be an Iray capable graphics card with significantly more then 4GB of ram on it that dose not run so hot that cooling it makes working with audio impossible.
In any case. I had done all them Iray benchmarks in Studio 4.9 and from habit I had done that last test run in Studio 4.8.0.59 by mistake. That mistake had me stumble onto something rather interesting to say the least. It took the GTX960 around 3005 seconds to run that to 2096 iterations in studio 4.8, redoing the render in Studio 4.9 had produced 2093 iterations in around 1848 seconds, around a third faster on the same card. Now that is an older version of Studio 4.9 mostly to keep the Iray benchmarks I've done consistent with each other, and there has been at least one public build update since 4.9.2.70 so render times may have changed since then. Also I've yet to do a 3DL test between them, mostly because of a lack of time, and I don't recall any major changes there except for non-render stuff (including some GUI optimizations that I wish there was a better way to speed up without changing the functionality of some stuff).
Time for some coffee and to ketch up on some things.
BTW, yes there was other changes in Iray between Studio 4.8 left, and Studio 4.9 right.
I'm sure Fred/Sabby provided updated shaders for FWSA Paloma, I just didn't bother to change anything to keep things consistent for render time testing.
OT. I think I can share the link now, I don't know if Nexus has the same 7-day exclusive that other channels have. In any case, from a former post on heatsinks, Andrew did a realy good job with that animation I think.
http://www.gamersnexus.net/guides/2603-new-gn-series-tldr-with-heatpipes-cpu-heatsink
Nothing much going on today, as I was just catching up on the past several years of Q&A vids by AgentJayZ, tho a different Jayz did just post something of interest. Apparently if you have more then two GTX10xx graphics cards for rendering in Iray, some gams just flat out crash the computer when you try to start the game.
http://www.youtube.com/watch?v=TCCEBfoUUPU
Now granted that was with three cards in SLI, tho I suspect it would make no difference at all if they were not in SLI, nor is it worth Jay's time to try a million other things. Three (or more) cards no longer works, and that is all there is to that. It also brings up some interesting doubts about VCA boxes with newer cards in them, even tho those are network connected things solely for rendering, the fact that they have more then two cards brings up some doubts with new generation cards.
I'm still around and kicking, I was just looking at a few things regarding computer cooling and taking a few notes from a different area that specializes in moving air. Yes, AgentJayZ does have some realy good presentations that covers a lot of stuff regarding jet engines.
The top two screen-caps are of a "bellmouth intake" that is for reducing 'Intake Depression' to allow a better flow of air into the engine. Without that, the air on the sides is turning around a sharp corner to get into the compressor, that slows down the air there, effectively reducing the diameter of the intake.
Now, I have lots of room in front of the Fan Baffle in my workstation to do whatever I want there. Tho I'm not quite willing just yet to start fussing with that foam that was put there back in 2005, as it is not the normal run of the mill weather stripping from the local hardware store. It's a tad more rare (automotive), and I don't think I have any of that left, so if it gets messy by pealing off layers it's game over. There is something else I can play with, hinted at by the two lower screen-caps in the first pic, intake Guide vanes.
The case I'm getting for my mother dose not have a lot of room in front of the intake fans, so I'm still considering how I'm going to cobble that together. So I was looking at just how thin a set of Guide vanes can be and still be affective.
Another thing I was considering was that Easter egg sitting there, It's the perfect shape for going on the center of the fan, It's just a matter of finding some that matches the diameter of the fan motor and attaching/suspending it "In Front Of" the fan (not on the fan).
So, I've been having a bit of fun trying to figure out what I'm going to do with mom's new computer case.
As for the stator blade set, I've already figured that out years ago.
http://www.zarcondeegrissom.org/ideas/016TurboCompFans.html
BTW, the very first thing that looks like wings on the intake of a jet engine, are actually 'Struts' that support the front bearing. Still, I think they help guide the air into the compressor blades at the correct angle given the shape of them (by design or accident), and some engines need a tad more with a second set of smaller blades that actually are called intake guide vanes.
After so many years of computers getting hotter, I find it odd that manufacturers have not attempted to use similar techniques That Actualy Work, with computer fans and fan mounts.
Hi again =) I'm not sure my youtube reply went anywhere (your comments were the first I've ever received there). So just in case - my dragon test vid took under a minute per frame on my laptop (or I wouldn't be rendering the animation). I was being lazy and ran the render to image series right from DS. Although the standalone, being newer, may well render a tad faster. I just didn't want to bother with checking if DS exports correct RIBs for animation etc.
As for V.Yudin's vids, I think he runs a crazy powerful developer rig, judging by how fast his IPR (interactive progressive render) updates everywhere. But the point is, the thin film alone is virtually free. Other shading will eat the bulk of the time.
Fun tidbit: half the problems with AoA's network is that very few things in it, if any at all, are "physically based". Not AoA's fault, it's just that the whole RSL thing in shader mixer can only be done the "oldschool" way. Which isn't exactly efficient for more complex stuff.
PBR - "physically based rendering" - doesn't stand for "the best in light transport calculations". It only means being energy-conserving and using "physically plausible" BRDFs (so that highlights that are bigger will be duller, for example). Most today's 3D games do PBR, real-time. Even though they can't do actual SSS or refraction.
Have you seen this video on cooling?
If you can get over the fact that he initially ran the $4000 card without any cooling!!!!!!!!!!
I've been trying to guess his profession.
Definitely not an engineer, my guess is physicist, the theoretical kind.
Prixat, with all due respect, that card was not run without any cooling (I Don't think he removed the heat sink), as the computer did have an intake fan in front of the cards blowing fresh air at them. That is how many rack servers are set up as well, with a row of fans blowing from front to back. The fault is on nVidia for expecting a card that pules more then 50 watts per slot to be able to stay cool without a J79 compressor cooling it.
That exhaust side duct he made looks rather cool. I have considered similar in the past, and sealing up a hole for wires plugged into the back of the computer to exit the exhaust fan duct makes it a difficult spot to put a fan (The Iray card is the first thing I've ever had that dose not have cables plugged into the back of it.). I had considered similar in the past to pull air threw the motherboard ATX plugs to help cool the CPU v-reg, I ended up just putting a baffle (Guide vane) on the intake fan to guide air down on the v-reg from the front of the computer (back in 2002 on a AthlonXP system before I had a camera). I started with cereal box cardboard and tape to figure out what shape was needed, then I used the cardboard to trace it out on more durable stuff.
The only thing I have of the beast was a bal-cam capture from the scrapping of the computer in 2005, and the quality is just horrendous. I never got a pic of the duct that went from the floppy drive opening straight to the Athalon XP 2500+ CPU cooler.
80% of the air that goes threw a jet engine, is for cooling the engine... Of course there is a better way then just using a J79 compressor to cool the graphics card
1, move the thing so your not blocking all the expansion slots. 2, reduce the power requirements of the thing. And 3, get a bigger heat sink.
duh.
Kettu, I need a cup of coffee, nothing bad, it's just an involved topic, lol. I did give the response a thumbs up when I read it, and it applies to so many different things (Opacity, ray-trace reflections, etc), that a simple youtube response can't completely cover. I do agree with the "If it takes more then X amount of time" then I kill it and find another way, I have to much to be doing to be waiting for my computer to get it's GUI out of it's back-end, lol.
I am also having another difficulty. The SATA power plugs going to the lower HDD racks have aluminum wires and are giving me some intermittent power difficulties. I tried re-crimping them and then soldering them to no avail. I'm going to have to find some SATA power pigtails with copper wire to remake that setup.
We bought some machines as a job lot that I had to rebuild. They were all different but a couple had this feature.
The round grill leads to the tube that feeds the CPU cooler directly.
You can also see where I taped over the various open meshes to re-institute some "directed flow" even though thats out of fashion these days.
VOOV! lol. Nice looking comp. yea, I had stuck some cardboard in the vents over the PCI slots to try similar, I had mixed results with that depending on the config of the baffles in front of the cards, and how much vent room was between each card.
Taping up the vents to prevent hot air from being sucked back in by the exhaust fans worked really well above...
A scoop to force air to go the full length of the cards before going to the back vent may also be 'OK' for intake fan only setups.
As for the duct leading straight to the CPU, it can work extremely well, provided there is nothing up against the side of the computer. I've done similar from the front, and kind of stuck with it over the years. Nothing beats dedicated air for the CPU.
Hay, it's Thursday. It's also starting to get a tad chilly outside, so it's a good time to think about warm thoughts. Going on the nVidia joke about enthusiasts sharing the sound samples of there watt-hog graphics cards. How about a flashback test run of a J79 from a former "Afterburner Thrust-day".
http://www.youtube.com/watch?v=YF-OkdUuA1k
I suspect that momentary negative EGT (Exhaust Gas Temperature) reading was a lose connection somewhere, it happens. Thankfully high end CUDA cards don't sound like that
, yet.
The beginnings of HPF (High Performance Fans), to replace the discontinued A2029 fans, Grrrrr, lol. Some stuff won't be here till after Monday, tho I am able to do some stuff now (your loss Thermaltake, sayonara).
I tried the Easter eggs I got back in 2005, and there just a tad bit to large for them to work on the Noctua fans. I may be able to trim that down, tho I'll need some other parts before I look into that.
Time for a test, that I only did on 80mm fans back in 2007/2008 about.
It involves some sowing thread, a dead PSU cassis, and power for the EUT (Equipment Under Test). You want the thread long enough that it's normal bends are close to the fan, and you can easily see what the air is doing on the exhaust of the fan.
And mounting the fan so the thread hangs straight down with the fan off is the bast way to do this. That way all the deflection in angle and rotation is what the fan is doing and not gravity. The angle will increase the more restricted the vents are in a computer, this is a No-Restricted air flow test.
Noctua NF-12 above, vs Delta FFB1212EHE below.
Looks similar to what I had seen back in 2007/2008 with the (Discontinued) External thermistor Smart Fans, between 20 and 40 degrees exhaust rotation depending on air flow restriction of the computer case. It looks like the same plastic coffee can will work fine for stator blades for these fans as well. Good to know.
I now need to figure out what the dip switches do on this thing. The only thing I can find on line is in 'Claw-marks' And I can't read 'Claw-marks' ( I don't even know what it is in, the translators can't figure it out).
Wow! what an impossibility. As mention last month, my bank gave me a new card, and so far changing stuff over to the new card has been rather painless. The domain registration that had been done threw GeoCities (remember them) for my website is now in such a tangled yahoo mess, that it is impossible to do something as simple as update the billing info. No, I am NOT going to purchase a second phone just to change a credit card being billed, and I have been done with multiple email accounts for over sixteen years (Both are required to get access to the billing part of Yahoo business now). Oy Vey.
I had been with Yahoo mail and GeoCities since 1996, this may be the last straw that forces me to go else where for email service. I moved the website to NYI as yahoo made the servers to slow for even simple HTML text pages, leaving just the domain registration with yahoo. I may contact NYI over the next month to migrate it all over to them, as yahoo is now impossible to continue using.
As if I don't already have enough to worry about, lol. I'm supposed to be at a Pow Wow today, I'll deal with this later.
What is this obsession with jets?
By the way I just subscribed to that youtube channel you linked to.
To paraphrase AgentJayZ, a jet engine is not a fashion statement or a branding for a company name. If it dose not actually make the engine run significantly better, it simply is not there. Jet engines are minimalist by design to an extent, because there is nothing on a jet engine that is not needed for something.
Just looking at the thumbs for AgentJayZ's vids, I can point out so much that computer fan makers should be doing instead of the wild (and some useless) designs that have come out the past sixteen years. And it comes down to how much energy are you putting into moving air threw a computer verses how much energy is wasted in turbulence and vortices (rotating exhaust air, not to be confused with "Wingtip Vortices").
The rotation of the air out of the fans I tested is quite visible in my former tests, and that is extra audible noise and reduced fresh air for cooling the computer. Not to say, the less efficient a fan is, the faster it must spin to keep up, and some are so bad they just cant do it at max RPM.
Not to say, some stuff on a jet engine is very Steam punk like, and I'm fond of steam punk like stuff of the functional kind.
The box of donor fans just arrived, so I can start making stator blade assemblies for the 120mm fans. Or more accurately "Exit Guide Vanes", tho not as extensive as these (I'm just using the shape of the coffee cans and the Easter eggs).
PS. I'm sure AgentJayZ appreciates the subscriptions to his channel. I know he has taught me so much about working with metal the past few months of watching his vids, and a lot of stuff that can be directly applied to modifying computer cases. Like why the engine above has double exit guide vanes instead of the older single blade design (hint, hint, computer manufacturers).
(EDIT) Because it is Monday (or was), and it is about that time of the year to find creative ways to remove white stuff, lol. A Blast from 2013.
http://www.youtube.com/watch?v=CVHuyr1x4Eg
There is a few parts of that engine test setup that I am making a facsimile of for computer fans, so it's a good reference as well.
That funnel is not exactly a bell mouth, tho it is enough to make a significant improvement over just mounting the fan directly on a flat computer case panel.
Bit of a Thursday morning update. It looks like Nicole is already on Bermuda, I hope everyone is hunkered down there. I'm already seeing sustained winds of 72km/h (about 50mph) at the airport (as of 12:08utc), so all the prep that was going to be done safely has been done at this point.
Keep your heads down, and Be safe y'all.
(EDIT, sustained winds 107km/h at about 13:40utc. About 66mph.)
EVGA Precision OCX is giving me a bit of grief. The fan curve is NOT being executed by the card at boot-up despite the power and temp levels running at there new settings.
And it is NOT starting up on windows boot despite the setting for it ticked. Grrrrr. Also the program is slower then the creative volume control to fire up when I manually click on it (in excess of a minute), so the card is just baking away that entire time till the program gets its GUI out of it's back-end.
The Turbo-fan mod is progressing. I initially measured the diameter of the central vain structure before I trimmed the Easter egg height down to match the diameter of the Noctua motors, now it dose not fit, oops.
I may just find a spice bottle of the correct diameter to put there as a spacer, as the egg plastic is rather explosive when being cut any way. It's one of a few options I'm considering as the computer case has yet to arrive to start work on the bell-mouth intakes. So that is where I am at this point, time for some morning coffee, and to look at some stuff.
Not much happening at this point, life is dragging me few other directions. I did play with the fans a bit more.
And I must say, I'm starting to like how it looks.
Nothing final, I'm still working on how it will all fit together. The Fractal Design computer case has yet to arrive (Must be held up in Swiss customs, lol), so I can't do much else for the intakes at this time. Chat later, I must run to carry speakers for a friend.
Just something wild, It's kind of counter-intuitive and against what you would normally think happens in a widening air path. Diffuser, or a "Divergent duct".
http://www.youtube.com/watch?v=uwh72hc-gWM
LOL It's not counter-intuitive to me but that's because I studied Aero engineering.
Here's everything I learned about air...
Air is much thicker than you think.
It always prefers to mix instead of flow.
Yea, fluid dynamics is an interesting field that I'm not sure GPU makers and PCI-SIG fully understand the ramifications of where things have gone the last decade. Something did arrive today, and at least it will not need to house such insanity for my mothers web browsing purposes.
Joke is from the add at the beginning of this vid, it's funny.
http://www.youtube.com/watch?v=o7srjqsEeIc
So, it looks like the front fan mounts do not support two 140mm fans, only a single 140mm in one of two possible locations, or two 120mm fans. That's kind of good for this, as I can set them up more like the B1B or TU-160 or Concorde intakes.
Over all the front looks good, with only a few minor adjustments it will be more then sufficient.
I haven't gotten much further then looking at that so far.
Over all I'm very pleased with the layout of the mATX case.
Wow, so according to a friend, the No-flow fans are based on submarine props
, to reduce cavitation noise!? 
...
A prop like that is designed to move a much more viscus fluid without a duct to contain the fluid flow... (Face palm). OK, forget about the containment duct, how fast must the air be going over a 120mm fan blade to have the air flow part from the back face of the blade and cause 'cavitation' vacuum voids in air... "Mach one" and no less. At lesser speeds the air is viscus enough to flow into the aria behind the blade as the blade moves. So, how fast must a 120mm fan spin to achieve Mach one at the tips of the blades?
(*1) Uh, I think the plastic the blades are made of will explode long before you can get a 120mm fan up to that kind of RPM, regardless what profile angle the fan blade are at. lol. I can assure you that 'cavitation' is not involved with computer fan noise, so what dose cause noise in computer fans, two things, transition tone noise and restricted flow turbulence (a stall).
Transition tone noise is easy to defeat, change the angle of the struts and stators so there is no high pressure along the entire length of the blades... like the strut inlet guide vanes on the Olympus 593 engine.
Fan makers have been doing that with the struts on fans going back before the IBM 8088 PC, so why not do that with the stator blades on high flow fans? (poke and prod at Delta and Noctua),
duh!
As for restricted flow turbulence (a stall, not cavitation), that is in the hands of computer case makers to make sure the intake vents and exhaust vents on a PC case are large enough to not restrict the flow threw the fans. Shallower pitch fan blades will produce less turbulence noise when the flow is restricted, however they have a lower top-end CFM unrestricted compared to steeper pitched blade fans.
Credit for fan blade turbulence reference pic... AutoDesk (and Wikicommons).
http://auworkshop.autodesk.com/library/sim-workflows-sust/metrics-and-basics-fluid-flow
As long as your not trying to cram a metric tone of air threw a single PCI slot, it is not that difficult to get a computer case fan to run quiet at a reasonable RPM and CFM for the computer. A bell mouth intake and a nose cone will also go a long way to reduce turbulence noise and increase air flow. Although, some fans are designed so poorly that they are always in airfoil stall above the motors minimum RPM, lol.
(EDIT 27Oct2016) *1, "4k RPM" is the max RPM of the Delta FFB1212EHE that I have here, It is not the max RPM of the entire Delta lineup of fans.
That "cavitation" thing cracked me up. Kinda like that piece of news I read today at a Russian email website: "Eta Carinae will destroy life on Earth".
Yea, we were both chuckling over the fan sales pitch wording, Everything about it was just wrong. As for 'Eta' doing something, well 'Something' always happens, lol.
If that is indeed Eta, then even if it makes a GRB, I don't think it will be headed this way. Unless there wrong about the jets traveling away from the poles of the star that we apparently are not looking down either pole of from here on earth, lol.
Besides, even IF there was a GRB headed this way, it's not worth worrying about, because nothing we do will change anything about it, lol. So carry on and enjoy life as it is.
(EDIT 21Oct2016) in separate news, ExoMars, congrats and sympathies. So what's the score now? something like Mars 27 Earth 28?
http://www.youtube.com/watch?v=ya05tVoEk8E
As for the fans, the Easter eggs are proving to be very brittle and difficult to trim down to size.
At least the guide vanes are going rather well.
Not much else to report so far.
So, yesterday I was fussing with the fans on the work bench when I inadvertently smacked the work light on the bench. Oops, After cleaning up all the bits of glass shards on the fan parts I took a closer look at the light (Before taking a pic of it). It's one of the more popular brands that I keep finding the bad ones that burn out within a year.
It looks like some of them don't have the LED segments flush against the heat sink core, and they don't come with any thermal compound at all in there. So, That explains the rapid failure of them, and it is sort of easy to fix with some white silicon goo. I now have a light without a cover that I must find something to do with. I'm not a big fan of live mains potential circuits just flapping in the breeze, lol. That light is not the first LED light I came across that didn't have a good thermal connection on the LED (a complete lack of thermal paste and sloppy mounting).
Tho this one had a few other difficulties, not to say a complete lack of light output that condemned it to sitting in a box unused for a few years now. That one was for "Regulated 12.0VDC" instead of 13.8V automotive power, so It is not usable at all for me here. That little bit of heat shrink also contains more then just a cable splice... Woo-a-a-a-ah!?
I don't think that little circuit thing can handle the amount of power the LED is using.
The circuit gets a tad hot and the LED is not bright enough for a "Flood light" as the unit was sold as (Sign spot light possibly, not an observation deck flood light). I've yet to see BigClive open up anything like that little circuit, so I have no idea what it was supposed to be before ending up in that light fixture. Enough of shotty LED lights, I just finished cleaning up all the glass shard bits from fussing with the two lights, on to other things.
Wireless, everything is going wireless, so why not wireless anti static wrist straps, lol.
http://www.youtube.com/watch?v=nvgW5iWXbts
ROFL. That's a good one 'Dave'.
(EDIT 24Oct2016) I just came across another light with the same PCB in it that BigClive did dissect...
http://www.youtube.com/watch?v=oKNh_uiK5wQ
Same guts and cheep construction, different housing.