Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Necere and wahaha360, how is the project coming along so far?
That's a bit of an oversimplification. You have to think about even the smallest things carefully, and there's a million little decisions that have to be made. How do I best balance cost vs. complexity on how this part is designed? How does doing something one way affect how this other part needs to change? Will doing it this way create any problems with parts fitting? What about installation? Is this going to be strong enough? Is this going to cause some problems for us down the line that will be difficult to correct? Who am I going to piss off by not supporting this or that feature? Is this component restriction to strict? And so on.
It's a difficult process if you want to do it properly, and yeah, it can be hard to muster up the motivation and focus to work on it by yourself. Especially so since the engineering side of things isn't something I particularly enjoy. I have no formal education as an engineer (or as a designer, for that matter). I do the technical detail work because there's no one else to do it, basically.
Apart from the difficulties of the design process, I'm not totally happy with the flexATX power supplies I tested. The likeliest candidate for this project is the 400W gold-rated flex by FSP, but even with some custom work they did on the fan curve for us, the noise profile isn't quite where I'd like it be. The overall sound level isn't bad at all, but the fan has as distinctive - though quiet - chattering or fluttering sound to it at idle, which changes to a mild buzziness under load. It reminds me of the sound of an AIO pump (though to be clear, not an air-in-the-loop sound). I assume it's caused by the ball bearings the fan uses, so we're seeing about getting a different fan to test in the unit, that will hopefully offer some improvement.
Flex atx too loud? Why not just use internal hplex psu? It makes no noise.
Flex atx too loud? Why not just use internal hplex psu? It makes no noise.
I'm all for flex if it doesn't sound like a jet engine taking offi don't want to create unnecessary panic towards Flex fan noise.
The overall fan noise is ok, we are just trying to get rid of, what we believe, the bearing noise on the fan (it's a ball bearing fan).
We are getting a new fan for the Flex, it uses FDB (fluid dynamic bearing) that you find in Noctua fans, we are hoping that will resolve the issue.
Waiting for it to arrive for testing.
The point is to get as much worked out before ever getting a prototype made. Invariably there's going to be things that can't be foreseen and will need to be changed once you have a prototype, but you want to reduce those as much as possible in the pre-prototype phase, or else you risk problems compounding. There's also things that are unpredictable, like the factory screwing up. Remember when Lian Li got the M1 files and messed up by leaving out the PCI bracket cutouts at the back edge, so I couldn't even install a card to test it without Dremeling it out myself? Or the ATX PSU bracket that was flipped from the way it's supposed to be? Finalize as much as you can in CAD first, then get a prototype made. Otherwise you're just wasting money and time.He is trying to do too much in prototype 1.
I have more modest goals for a prototype, because I know there will be several protoype revisions.
Noctua uses their own maglev/fluid bearing that they call SSO. FDB is a sleeve-type fluid bearing, which is supposed to have a longer life than regular sleeve bearings, but less noise than ball bearings.We are getting a new fan for the Flex, it uses FDB (fluid dynamic bearing) that you find in Noctua fans, we are hoping that will resolve the issue.
The point is to get as much worked out before ever getting a prototype made. Invariably there's going to be things that can't be foreseen and will need to be changed once you have a prototype, but you want to reduce those as much as possible in the pre-prototype phase, or else you risk problems compounding. There's also things that are unpredictable, like the factory screwing up. Remember when Lian Li got the M1 files and messed up by leaving out the PCI bracket cutouts at the back edge, so I couldn't even install a card to test it without Dremeling it out myself? Or the ATX PSU bracket that was flipped from the way it's supposed to be? Finalize as much as you can in CAD first, then get a prototype made. Otherwise you're just wasting money and time.
Jesus. I haven't wanted to say anything, because it's totally not my place, but this is getting a little silly.
Guys? I suggest that you both take a break. Walk away for a day or two, chill off, and think about the reasons that you're working together in the first place.
Right now you're bickering worse than a high-school band, and that's neither going to do good things for your working relationship, nor let this project ever see the light of day.
Jesus. I haven't wanted to say anything, because it's totally not my place, but this is getting a little silly.
Guys? I suggest that you both take a break. Walk away for a day or two, chill off, and think about the reasons that you're working together in the first place.
Right now you're bickering worse than a high-school band, and that's neither going to do good things for your working relationship, nor let this project ever see the light of day.
That's a bit of an oversimplification. You have to think about even the smallest things carefully, and there's a million little decisions that have to be made. How do I best balance cost vs. complexity on how this part is designed? How does doing something one way affect how this other part needs to change? Will doing it this way create any problems with parts fitting? What about installation? Is this going to be strong enough? Is this going to cause some problems for us down the line that will be difficult to correct? Who am I going to piss off by not supporting this or that feature? Is this component restriction to strict? And so on.
It's a difficult process if you want to do it properly, and yeah, it can be hard to muster up the motivation and focus to work on it by yourself. Especially so since the engineering side of things isn't something I particularly enjoy. I have no formal education as an engineer (or as a designer, for that matter). I do the technical detail work because there's no one else to do it, basically.
Apart from the difficulties of the design process, I'm not totally happy with the flexATX power supplies I tested. The likeliest candidate for this project is the 400W gold-rated flex by FSP, but even with some custom work they did on the fan curve for us, the noise profile isn't quite where I'd like it be. The overall sound level isn't bad at all, but the fan has as distinctive - though quiet - chattering or fluttering sound to it at idle, which changes to a mild buzziness under load. It reminds me of the sound of an AIO pump (though to be clear, not an air-in-the-loop sound). I assume it's caused by the ball bearings the fan uses, so we're seeing about getting a different fan to test in the unit, that will hopefully offer some improvement.
He is trying to do too much in prototype 1.
I have more modest goals for a prototype, because I know there will be several protoype revisions.
The point is to get as much worked out before ever getting a prototype made. Invariably there's going to be things that can't be foreseen and will need to be changed once you have a prototype, but you want to reduce those as much as possible in the pre-prototype phase, or else you risk problems compounding. There's also things that are unpredictable, like the factory screwing up. Remember when Lian Li got the M1 files and messed up by leaving out the PCI bracket cutouts at the back edge, so I couldn't even install a card to test it without Dremeling it out myself? Or the ATX PSU bracket that was flipped from the way it's supposed to be? Finalize as much as you can in CAD first, then get a prototype made. Otherwise you're just wasting money and time.
Noctua uses their own maglev/fluid bearing that they call SSO. FDB is a sleeve-type fluid bearing, which is supposed to have a longer life than regular sleeve bearings, but less noise than ball bearings.
Anyway, in FSP's most recent email they're saying they've tested the FDB fans we've asked about, and they found them to still have problems. Instead, they're going to send us a Delta fan (original and FDBs are Protechnic) to test, which seems to still be ball bearing.
...Also... I'm pretty sure... wahaha360 is bored to death (and wants to make some money) and has been for the past few months (if you check, some time ago he would barely write anything because he was sourcing some new manufacturers).
Necere, isn't any brooming to be done? I see some dust on the corner over there... I'm sure wahaha360 can get stuff going xD at least he won't have as much time to be pestering you.
...in my opinion time is more expensive than prototyping cost
Hahaha, I am bored and looking for excitement.
On a more serious note, this is nth time multiple manufacturers have reached out to me about what we want to do, I'm just tired of telling people we are not ready.
Well, at least you know that when you become ready, there are companies out there more than willing to work for you.
BTW, I thought you mentioned some time ago about... sandblasted panels for the M1? How is that going? You plan to change suppliers for the M1? I'm bored of my case (Phanteks Evolv ATX), and it is simply too big, and I want some new hardware... so, I'll have to buy a new case. I'm pretty sure you are at least 6 months away for your next release, but sand-blasted panels for the M1 would do it for me (for the time being). Specially since I sold my V1 because I thought that, with enough time, the side panels might be too deformed to be used properly. Heck I can tell you that they weren't as when I bought it, simply because of the amount of times I dissassembled the computer.
Hahaha, I am bored
On a more serious note, this is X number of time mfgs have asked me for update, I'm tired of telling them I don't know.
I always liked making money, these projects didn't changed me I also live in NYC, it's just expensive here, you really have to be rich to not think about money living here.
NYC! Me too. Can I just walk over and pick one up?
BTW. Why don't you make a prototype with a 3D-printer? By this stage either you or wahaha360 have to know somebody who could make a prototype without much trouble. At least you will have something to mount your stuff and have a more general idea of what it will look like. You can't make thermal tests on something made of a completely different material but you can test how mounting would be without spending a thousand dollars and waiting weeks to get a prototype.
Haha, I don't have the case, believe me I wished we were in production now.
Ever since we had access to MacBook like finishes, I been itching for a hackintosh build and maybe VESA mount this case on the back of my 32" monitor
We had a meeting and agreed that I need progress for my sanity, so I will work on design details to make prototype 1. I just want to make sure there are no surprises / problems (which we had with LRPC).
So I did a bit more testing and it turns out the noise that was bothering me in the FSP 400W Flex PSU isn't so much the fan bearings after all, but some kind of coil buzz/clatter that manifests during higher loads. Here's a recording I made of the sound:
Note that the 3DMark testing (and therefore the coil buzz) doesn't start until 1:18. I manually stopped the PSU fan periodically so we can be 100% sure it isn't the source of the sound. At idle with the fan stopped, the PSU is totally silent. At load, apart from the coil buzz, the sound of the fan isn't bad at all, though it does still have a bit of bearing noise (which doesn't come through that well in the recording). Remember this is with the microphone less than an inch from the fan intake.
You can hear it for yourself. I normally think of coil whine as a high pitched noise, like in this video. The sound coming from this PSU sounds more like a tiny tractor is in there. But it behaves similarly to the more typical coil whine, changing depending on GPU load. For example, I was testing it with Dishonored 2, and I noticed it was worse when I looked at the water in game.Would you call it coil whine? There's been reports of some coil whine from S4 Mini builds. I think in particular it's when running stress tests. Seems like these small PSUs are more susceptible to coil whine than their larger counterparts.
MMmmm...looking good - I'd thought you and Necere doing away with the optical drive....
HDPLEX 300W Hi-Fi DC-ATX Power-Supply:
HDPLEX Internal 300W AC-DC Adapter:
I'm just going to leave this here. Also, I prefer circular vent holes instead of square ones, just like Valve's Steam Machine...
PLEASE! ANYTHING BUT SQUARES!
I wonder if a 400W HDPlex AC-DC brick is possible without a fan and within reasonable dimensions.
I'm not in favor of separate AC-DC/DC-DC adapters/boards, personally. They have lower wattage limits, cost more for the wattage, aren't typically designed for "peaky" power draw of a desktop system, and aren't standardized. We don't want to be in the PSU business (nor is it very feasible for us), and I don't want to be reliant on a single vendor for a proprietary PSU. Consequently, we need to design for a standard form-factor, like FlexATX or SFX.Have you guys looked into VHF power conversion, as provided by e.g. FINsix? They are selling a tiny 65W laptop charger, but I'm not sure how ready/interested they would be in a higher power requirement application (and you'd still need something like the HDPlex DC-ATX board).
I'm not in favor of separate AC-DC/DC-DC adapters/boards, personally. They have lower wattage limits, cost more for the wattage, aren't typically designed for "peaky" power draw of a desktop system, and aren't standardized. We don't want to be in the PSU business (nor is it very feasible for us), and I don't want to be reliant on a single vendor for a proprietary PSU. Consequently, we need to design for a standard form-factor, like FlexATX or SFX.
I think that's wise. And if anyone wants to use HDPlex in this case, it's a very easy mod with some velcro/double sided tape.
1) I don't know a lot about airflow/aerodymanics, but if I was to set the intake fans to a low RPM, and the GPU ramped up during play, or I was using a dual-fan card that fit, is it possible the card would start intaking from the exhaust vent if it's not getting enough air? Especially with the Flex ATX PSU eating up airflow? Or is that not possible with the exhaust vent being the only exhaust, i.e. if the card is not getting enough from the intake fans, it would just recirculate some of its own air? I'm having trouble imagining it.
I2) Is the ODD bay back for good? Or do you plan on offering both versions? Would it be possible to use it as an additional 2.5" drive mounting spot? Generally do you plan on having additional SSD mounting locations elsewhere in the case or is there no room?
I3) I can't tell from the render, but would the slim fans go on top of the bracket or under it? I had a crazy thought of moving the PSU to under the bracket, taking its cover off and 40mm fan out and using a thicker 120mm fan to cool it directly from the bracket to eliminate noise. Then you'd have room for a full-length card ready to receive all of that warm air And partially why I asked the question about additional 2.5" mounting points. How would you rate that on the insane mod scale?
I4) Do you plan on including an option for mounting a 3.5" drive in place of the graphics card for people who want a non-gaming machine/NAS? Or is that such a niche use case that it's not worth it.
If the GPU really ramps up I think the main thing that might happen is that it will recirculate some of its own exhaust. It can't really pull air in through the exhaust vents, since the air coming into the case needs to go somewhere, and the only other exit is the small exhaust vents on the rear of the GPU bracket. The GPU would need to be ramped up pretty high for it to be moving more air through those small vents than is coming into the case through the two 120mm intakes.1) I don't know a lot about airflow/aerodymanics, but if I was to set the intake fans to a low RPM, and the GPU ramped up during play, or I was using a dual-fan card that fit, is it possible the card would start intaking from the exhaust vent if it's not getting enough air? Especially with the Flex ATX PSU eating up airflow? Or is that not possible with the exhaust vent being the only exhaust, i.e. if the card is not getting enough from the intake fans, it would just recirculate some of its own air? I'm having trouble imagining it.
If the GPU really ramps up I think the main thing that might happen is that it will recirculate some of its own exhaust. It can't really pull air in through the exhaust vents, since the air coming into the case needs to go somewhere, and the only other exit is the small exhaust vents on the rear of the GPU bracket. The GPU would need to be ramped up pretty high for it to be moving more air through those small vents than is coming into the case through the two 120mm intakes.
I'm not sure how useful that would be. You'd see smoke going in and coming out, but what happens inside would largely be obscured by the components and chassis.Will Ncase be doing smoke tests to confirm the air flows as intended? I would love to see some photos or video of that.
I'm not sure how useful that would be. You'd see smoke going in and coming out, but what happens inside would largely be obscured by the components and chassis.