Lads talking beer is never a derailment, unless we were supposed to be driving the train?
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
The next NCASE project: a Steam Machine-style case (indeterminate)
- Thread starter Necere
- Start date
Lads talking beer is never a derailment, unless we were supposed to be driving the train?
A regular axial fan will usually have to work a bit harder when it has to suck air directly through something (side panel hole pattern and/or air filter), but on the other hand it will suck in ambient temperature air, not pre-heated by any other components. I wouldn't worry about the performance.
I do worry a bit about noise though, but it shouldn't be too bad. And there's not much to be done about it anyway, with the limited space.
Hell no! If you have them at the back, the signal has to go through a lot of metal - not good.
That would spoil the look.
I beg to differ. There are fanless cases which manage it - they use heatpipes and fins on the back. You just have to do a bit of manual adjustment when you build the PC.
I thought external 600W PSUs were available.
Unless the signal is very weak (like most or all of Logitech's wireless stuff is), it shouldn't really be a problem.
Would it really be worse than having four connected devices on the front of the case?
Making this case fanless would require a lot of changes, and the size would have to be increased substantially to fit the fins. But I'm not sure if that's what you mean?
As far as I know, Dell's 330W power brick is the most powerful out there, but not sure.
D
Deleted member 276776
Guest
Yeah, my bad. Fixed.
D
Deleted member 276776
Guest
Aluminum shouldn't cause any issues. We're talking 3-6mm max of aluminum when your wireless round doesn't have any problems covering most peoples' entire homes. Besides the signal isn't weak at all.
Pfffft nope. If only.
Unless the signal is very weak (like most or all of Logitech's wireless stuff is), it shouldn't really be a problem.
Would it really be worse than having four connected devices on the front of the case?
As far as I know, Dell's 330W power brick is the most powerful out there, but not sure.
This. The signal is plenty strong. If you really did care about latency and whatnot you would have wired equipment anyways.
I thin a single 50-pin header on your mobo is only capable of two USB 3.0 ports anyways. The rest would be USB 2.0 which, although not a problem for mice and keyboards, might be for other stuff.
I believe it is. FSP or Lite-On might have 19v ones like that too w/ the 400W peak.
energizerfellow
n00b
- Joined
- Aug 27, 2014
- Messages
- 42
That's what the USB hub built into your monitor is for, along with Logitech's Unifying Receiver. All the devices you mention are USB 2.0, so compatibility shouldn't be a problem. As has been mentioned, signal strength shouldn't be an issue.
That and all wireless headsets are evil. Stay far, far away.
No problem. I was a bit confused when I first saw that, so I thought I should mention it.
Is there a specific reason for the flexible riser as apposed to a rigid riser? I would think a rigid riser would give more structural support for a graphics card, less cost, and less risk of compatibility issues that others have noticed with flexible risers.
1o57
Gawd
- Joined
- May 17, 2012
- Messages
- 894
I bought an M1, and I'd buy this.
Bobalias_LeShay
Limp Gawd
- Joined
- Sep 27, 2013
- Messages
- 287
This isn't quite correct. 3mm of aluminum is more than enough to completely nuke a signal to nothing in the 2.4GHz ISM band. The reason you can still get a signal when there's metal around is because there are more paths for it to take than just straight through the case. Also there are holes and passive radiators it can sneak through/along. Your conclusion is correct (signal should be fine), but the reasoning isn't (3mm of Al won't kill the signal).
Personally, the cheap Logitech keyboards I've used have been so terrible that a receiver 10 feet away won't always work. Their higher end models may be better, but I'd avoid the cheap ones unless you're sitting 3 feet away from the tower. Though that may have something to do with using them in big apartment buildings with interference all over.
I gave some reasons earlier in the thread. It allows more design flexibility with regard to positioning the GPU, rather than being limited to 1U or 2U-tall risers. This can be useful to maximize the available space. It also allows for CPU coolers that hang over the PCIe slot. Support for the GPU shouldn't be negatively affected in this design, since the riser is attached to the chassis. Cost and reliability are concerns with a flexible riser, it's true, and it could still change to a solid riser during the course of development.
What a possible no-ODD version would look like:
I actually really like the button and ports centered like this better. Even if you don't do an ODD version, can you center these?
Considering it seems to be a rather unpopular option, it's doubtful. Centering the power/USB also limits GPU length to 10.8", so it's a bit too much of a disadvantage without a corresponding benefit if ODD support were excluded.
Are black feet an option?
Is asking for 120mm fan mounting holes centered over the motherboard out of the question?
Do you suspect pricing to be similar to the M1?
Will you be ordering a prototype to test again before production? If so will there be another fund campaign to pay for said prototype?
Is asking for 120mm fan mounting holes centered over the motherboard out of the question?
Do you suspect pricing to be similar to the M1?
Will you be ordering a prototype to test again before production? If so will there be another fund campaign to pay for said prototype?
shakes head at posts like this. get a $5 USB hub, or make him redesign the case. I know which I'd choose
Possibly, though I'm not sure the 35x13mm in the renders are available in black.
No plans for it.
It's possible it will be more, mostly due to the inclusion of the riser. We won't know until we have a more finalized design that we can get a cost quote on, though.
Yes, prototype(s) will be built and tested. Soliciting funding for that purpose won't be necessary.
To be honest redesigning the case for 4 USB ports on the front shouldn't be too hard considering there are already two in the current design. And usb hubs are ugly.
But for his needs you could just put the receivers on the back.
Exactly. Why the hell would you want all of these connectors in the front? Besides, Logitech has things like unifying receivers that can connect your keyboard, mouse and headset in a single receiver.
It would have to be two usb 3.0 and two usb 2.0 (like e.g. Antec P280), which apart from looking a bit crowded would also likely mean two blue and two black connectors. Not very pretty on a good looking case like this.
The front panel should be as clean as possible. I would actually prefer a completely blank front panel with only a power button. Havent used a usb drive in ages.
But as with the odd issue its really obvious that people have very different habits in reg to storage. To get this project off its feet it needs to cater to as wide an audince as possible.
by the way.
a couple of pages back i asked if it would impact more than the front panel holes to change to keep the power/usb on top for the no odd version and elsewhere on the odd version so that we who opt for no odd wouldnt have to do the gpu-length trade off. Does anyone know? Necere?
But as with the odd issue its really obvious that people have very different habits in reg to storage. To get this project off its feet it needs to cater to as wide an audince as possible.
by the way.
a couple of pages back i asked if it would impact more than the front panel holes to change to keep the power/usb on top for the no odd version and elsewhere on the odd version so that we who opt for no odd wouldnt have to do the gpu-length trade off. Does anyone know? Necere?
That's an interesting question! It should impact both the front panel and the panel piece behind it, but wouldn't it be possible to have two mounting locations (I'm guessing screw holes?) for the power + usb pcbs, making it simple to just manually move them to the ODD space for us who don't want it (which is how I interpret your question)?
Then, the ODD front panel would have power + usb on the left/bottom side (like in design variant 3 or 4), and the no-ODD panel would have power + usb on the right/top side (like in the original design).
That would be a win/win!
Last edited:
Not everyone uses or wants to be forced to use Logitech products. And receivers at the back of the case are a bad idea.
Get a USB extension cord/hub from the back and wire it to a discrete location where receivers are hidden.
I don't see the point in buying such a clean and sleek case if you're just going to stick a load of fugly looking USBs in the front.
Having two different front panel assemblies means having two different production runs for the front panel. This means quite a cost increase.
When the placement is kept the same, the only difference is that a batch can skip the ODD-punch stage to produce no-ODD panels. If the panels have different layouts, then the no-ODD panels and ODD-panels share only a common base material. They need to go to different punching machines (or require the machine to be reconfigured) to punch the front I/O in different locations, need two different validation setups (to confirm the punched I/O holes are sized and positioned within spec), need two different assembly plans for installing the front panel PCB, need two different front panel PCB validation jigs (assuming validation occurs post-install to avoid missing installation damage), etc.
When the placement is kept the same, the only difference is that a batch can skip the ODD-punch stage to produce no-ODD panels. If the panels have different layouts, then the no-ODD panels and ODD-panels share only a common base material. They need to go to different punching machines (or require the machine to be reconfigured) to punch the front I/O in different locations, need two different validation setups (to confirm the punched I/O holes are sized and positioned within spec), need two different assembly plans for installing the front panel PCB, need two different front panel PCB validation jigs (assuming validation occurs post-install to avoid missing installation damage), etc.
Allanitomwesh
Weaksauce
- Joined
- Jun 28, 2014
- Messages
- 74
It would put the slot super close to the edge of the case.
In the image,what would happen if you reversed ODD and I/O to have the ODD on the bottom and I/O up top? Is that even possible?
Yes. That was the idea. Of course, then there is this:
Yeah. You're probably right. I suspected something like this. It would put a lot of discussions to rest though.
It's not possible in the current design. The GPU fills the space in the bottom, so you would have to choose between GPU and ODD.
Had a thought. What if.....
The front panel is pre cut-out a perfectly symmetrical sheet of aluminium right?
So. If ODD version has I/O cutout in bottom, and no-ODD has them on top, then it is just a matter of flipping it up-side-down before the "fastening things" gets welded on. This means that they cut all front panels the same.
This might have a really logical thing stopping it as well. If so, I'll lay this to rest. Promise
Edit: The frame behind the front panel could be something like this (assuming i have any idea how this is put together, which i don't):
Last edited:
That's an excellent idea, I really hope it can be done!
Had a thought. What if.....
The front panel is pre cut-out a perfectly symmetrical sheet of aluminium right?
So. If ODD version has I/O cutout in bottom, and no-ODD has them on top, then it is just a matter of flipping it up-side-down before the "fastening things" gets welded on. This means that they cut all front panels the same.
This might have a really logical thing stopping it as well. If so, I'll lay this to rest. Promise
Edit: The frame behind the front panel could be something like this (assuming i have any idea how this is put together, which i don't):
Ah. Worth looking at atleast i think. If one goes like that, one could also make room for odd on bottom of the case too?
Do you have a time frame in mind on how long you will entertain changes to the design before you order a prototype? Everything thus far seems to be fairly cosmetic except maybe trying to fit an ODD in it. Because obviously this could go on forever, i mean look at the original M1 post, people still post requests for something to be changed years later.
To me at least, all the major design hurdles seem to be complete to make the case work within the design goals. Are there any fundamental issues you are still working on we aren't aware of yet?
To me at least, all the major design hurdles seem to be complete to make the case work within the design goals. Are there any fundamental issues you are still working on we aren't aware of yet?
The optical disc drive looks like it could fit in the bottom right corner and be placed under the PSU. Slot height is 13.2mm and the psu is 63.5. The psu width is 125mm and the optical length is 127mm. If the optical drive was placed there the length of the GPU wouldn't need to be changed. I understand that might be a huge change and completely out of line of the design but I thought I'd at least make the suggestion and hear your thoughts Necere.
D
Deleted member 276776
Guest
The signal should be able to make it through. A 2.4GHz signal has a wavelength of 0.125m so it should travel through 3-6mm of aluminum without too much issue (although there is other interference so yes, the signal will be weaker). But yes, the alternate pathways of the signal will still give you a pretty strong signal. 5GHz shouldn't pose too much of a problem either because you still have a wavelength of ~0.06m wavelengths which is still 6 cm of material max before the wave can no longer pass through the material and reflect off of it.
But if the waves do interact w/ the aluminum, yes it wouldn't be able to pass through due to basic boundary behavior.
AthlonXP
Fully [H]
- Joined
- Oct 14, 2001
- Messages
- 20,591
Ok so shut up and take my money! I really need something like this in my living room and this is perfect!
Bobalias_LeShay
Limp Gawd
- Joined
- Sep 27, 2013
- Messages
- 287
This isn't correct. You're right that the relevant calculation is a function of frequency, but it's also a function of conductance (and relative permeability, but that's 1 here). The skin depth of aluminum at 2.4GHz is on the order of 2-3 micrometers. You can make a pretty effective Faraday cage out of aluminum foil, try it sometime!
I think you're talking about impedance mismatches, but that's not the main issue here.
D
Deleted member 276776
Guest
That was the only part I was a bit unsure about b/c of the fact that aluminum is a good conductor.
But otherwise that's it. It will travel "around" via other paths which is good enough.
Bobalias_LeShay
Limp Gawd
- Joined
- Sep 27, 2013
- Messages
- 287
That's certainly true. If changing the transceiver from the front to the back of the case makes someone lose signal, it was a pretty terrible transceiver to begin with.
Yeah, I was thinking about something like that. I'll have to get into the details to find out if there's some dealbreaker there or not.
Mostly detail work at this point. But therein lies the devil
It doesn't fit because this is how the width adds up: 1.5mm (side panel) + 3mm (filter) + 63.5mm (PSU) + 1.5mm (motherboard tray) + 1.5mm (chassis front flange) + 1.5mm (side panel) = 72.5mm
Also there's not enough depth: 145mm internally between the chassis front and motherboard, while the slim ODD is 130mm but the slim SATA connector needs quite a bit more than 15mm.