NCASE M1: a crowdfunded Mini-ITX case (updates in first post)

Glad to see how well this all seems too bee turning out. Was just wondering about an option of getting 2 of the 2 x 3.5 cages. Also Lian Li never sends as many of the antivibration bits as is needed. Would be awesome case for a home server. 5 x 3.5 and a 2.5 in one super compact case just awesome. Keep up the good work gents.
 
WiSK said:
GTX680 draws less than 350W from the wall with an overclocked i7 @ 4.8GHz. Synomenon is using a CPU with 45W TDP, should be drawing under 300W from the wall.
http://www.hardocp.com/article/2012/03/22/nvidia_kepler_gpu_geforce_gtx_680_video_card_review/13
Click to expand...

meh, i think running a 680 on a 350W psu is just asking for trouble and i would never recommend it, even if the math says it should technically be possible. there are 450W psus the exact same size, and for not much more, so i would just stick with one of them.

either way synomenon, ditch that 3770t, its a dumb processor. they basically just charge you extra for an underclocked (slower) 3770k. just get a 3770k and a 450W psu and you can OC the crap out of your processor for basically the same cost.
 
ekuest said:
meh, i think running a 680 on a 350W psu is just asking for trouble and i would never recommend it, even if the math says it should technically be possible. there are 450W psus the exact same size, and for not much more, so i would just stick with one of them.

either way synomenon, ditch that 3770t, its a dumb processor. they basically just charge you extra for an underclocked (slower) 3770k. just get a 3770k and a 450W psu and you can OC the crap out of your processor for basically the same cost.
Click to expand...

I wouldn't say a 3770T is a dumb CPU. It has it's uses, maybe not for a gaming desktop, but for a server/NAS/HTPC it kinda kicks ass, and has a few features a K variant doesn't: http://ark.intel.com/compare/65525,65523
 
Diverge said:
I wouldn't say a 3770T is a dumb CPU. It has it's uses, maybe not for a gaming desktop, but for a server/NAS/HTPC it kinda kicks ass, and has a few features a K variant doesn't: http://ark.intel.com/compare/65525,65523
Click to expand...

i admit i dont know a whole lot about the "t" series, but why would you need an i7 for a server or htpc? also, i was assuming that he was only planning on the t in order to lower his power usage in order to use an sfx psu and/or save money on his power bill, in which case the same thing can be done for less with a k series, but also leave the option for overclocking if desired. could be that he needed some other feature of it.
 
ekuest said:
i admit i dont know a whole lot about the "t" series, but why would you need an i7 for a server or htpc? also, i was assuming that he was only planning on the t in order to lower his power usage in order to use an sfx psu and/or save money on his power bill, in which case the same thing can be done for less with a k series, but also leave the option for overclocking if desired. could be that he needed some other feature of it.
Click to expand...

Well I built mine because my Synology NAS wasn't powerful enough for Plex trancoding of HD movies. It was using all the CPU of the NAS. There's also some other cool features of that CPU, like ntel® vPro Technology, which is basically remote desktop to you PC at the hardware level. Can reboot a PC, change stuff in the bios, ect, all remotely.
 
M4rk said:
Looks like that case used to cost $400 before markdowns. When that case debuted 5+ years ago, $400 was probably worth more than $400 nowadays (material costs), so probably around $500.
Click to expand...

Some versions went for $649 a few years ago.
It would be a $600-800 case if still made. The TJ11 and Caselabs are an example of that.
 
folks, we definitley need a "short" & silent atx power supply.
there's no such thing on the market. for me this would be a major showstopper.
 
ekuest said:
meh, i think running a 680 on a 350W psu is just asking for trouble and i would never recommend it, even if the math says it should technically be possible.
Click to expand...

I'm not sure if you understood how [H]ardOCP does their testing, but they don't rely on "the math", they plug it into a multimeter and measure the actual power draw. Whatever your gut feeling tells you, put it aside and learn to trust the facts instead :)
 
WiSK said:
I'm not sure if you understood how [H]ardOCP does their testing, but they don't rely on "the math", they plug it into a multimeter and measure the actual power draw. Whatever your gut feeling tells you, put it aside and learn to trust the facts instead :)
Click to expand...

well for one they tested the rig gaming, not stress testing. they also did break 350 in one game. i wouldnt want a psu to ever go to 100% of rated anyway just because of noise and stress on the components. last i checked, it is recommended that you use about 75-85% of your rated wattage for longevity and headroom in case of power spikes or a bad cap or whatever else might go wrong. i may be wrong, but even if a computer is expected to use 330W, i think most psu experts will recommend against getting a 350W psu. one number is less than the other, yes, but there is more to deciding on which to get than that. i would heartily recommend paying $30 more to have 120W of headroom (with a quiet psu and not pushing the psu to the very limits just playing metro 2033) than save that $30 and have 20W headroom, and my psu screaming every time i fire up a game, my whole case heating up because the psu is stressed, and the knowledge that the thing could blow if i plug in too many usb peripherals. he could think "oh well metro 2033 only pushed that rig to 358W and im using a lower TDP proc so i will be fine" and then he fires up crysis 3 without remembering to unplug his external hdd and boom dead psu, and probably a motherboard or cpu with it.

i kept it simple before because i didnt want to have to write all that, but i think its irresponsible to say "stick with that one because the number is smaller" when there's so much more to take into account. it has nothing to do with me distrusting [H] and everything to do with seeing hundreds of computer problems on these forums, and knowing that so many of them resulted from psu issues.
 
In regards to the issues being raised with purchasing hardware with campaign funds (announcement sent to supporters): I don't see anything wrong with loading the case with the latest and greatest for testing and photo shoots. But then keeping all said hardware or possibly reselling it down the road for profit seems wrong (not say they will do either).

It may be proper to do a charity sale of a complete system, loaded with hardware and using one of the production run cases from the 2nd campaign to follow. Or sell individual components off for charity. It would be a great way to give a few grand to an organization that needs it and get the word out about NCASE and the M1.

Or even another option is to give them to [H]ard|Forum for use as contest prizes (not sure about legality/liability merit there). After all, the [H] is a big part of this cases development.

Of course, if you take the more conservative route and go with things like the 480 as opposed to the GTX690, they will be a surplus of funds from only 2 prototypes that can be applied to the production run purchase order. You could tack on another prototype for a total of 3 as well without exhausting all your funds.
 
I still think it would be much more cost effective to purchase electrically defective (mechanically sound) hardware from various forums (at least for the bulk of the fitting; obviously, you're still going to need a fully working setup that is a real space heater to test the thermals).

At least AMD processors are cheaper (and produce a crapton more heat) :p a 1090T can turn into a monster once you start cranking the voltage (or $80 for a new phenom ii x4 965 BE @ 140 watts stock :eek: )
 
I've also thought it would be fascinating to have FLIR images to validate thermal characteristics. Get somebody you know in the fire department and see if they'll lend some time on it to you. Better yet, bring the complete system into the firehouse and set it up for testing. My boys here in the fire services would welcome something different in the house, especially knowing most of them are avid gamers between runs.

And also believe you can rent the FLIR imagers and cameras now.
 
I don't mind them using the funds to purchase testing hardware as long as the two prototypes are paid for and they leave some reserve for a possible updated prototype. You don't just make a prototype and say "Welp, it pretty gud. Make'em." It needs to be tested thoroughly.
If they want to give the hardware away, raffle it, or sell it later on thats fine too.


Machupo said:
I still think it would be much more cost effective to purchase electrically defective (mechanically sound) hardware from various forums (at least for the bulk of the fitting; obviously, you're still going to need a fully working setup that is a real space heater to test the thermals).

At least AMD processors are cheaper (and produce a crapton more heat) :p a 1090T can turn into a monster once you start cranking the voltage (or $80 for a new phenom ii x4 965 BE @ 140 watts stock :eek: )
Click to expand...

If you are solely concerned with thermals your best bet is probably the ASRock FM2A85X-ITX and an overclocked AMD A10-5800K w/ a high-end, last-gen video card. It will put out more heat and draw more power than any current Intel mini-ITX system. There are no AM3+ ITX boards and if you can find an AM3 board it will be limited to a 95w TPD and probably not equipped with an x8/x16 slot. Then again I'm not sure what the cooling options are for FM2, I assume its the same as AM3+.
 
ekuest said:
well for one they tested the rig gaming, not stress testing. they also did break 350 in one game.
Click to expand...
That's 356W from the wall, so internally under 320W. And remember with the i7 @ 4.8Ghz as well, which has a TDP at stock of 95W while his low draw -T chip is well documented as drawing less that the rated 45W. So we have total system use under 270W.

ekuest said:
i wouldnt want a psu to ever go to 100% of rated anyway just because of noise and stress on the components. last i checked, it is recommended that you use about 75-85% of your rated wattage for longevity and headroom in case of power spikes or a bad cap or whatever else might go wrong.
Click to expand...

350 * 85% = 297W. So we're well inside your recommendation.

When you talk about longevity. If someone is running the CPU at full load 24/7 then I would agree, keep it under 75%. But Synomenon's PC is not doing BOINC or folding@home or something like that.

ekuest said:
i may be wrong, but even if a computer is expected to use 330W, i think most psu experts will recommend against getting a 350W psu. one number is less than the other, yes, but there is more to deciding on which to get than that.
Click to expand...

But it's not using 330W, which is why I linked the review to you. But I think you mistook the from-the-wall measurements for actual output consumption.

ekuest said:
i would heartily recommend paying $30 more to have 120W of headroom (with a quiet psu and not pushing the psu to the very limits just playing metro 2033) than save that $30 and have 20W headroom, and my psu screaming every time i fire up a game, my whole case heating up because the psu is stressed, ...
Click to expand...

The noise is exactly why I recommended the BeQuiet instead of the ST45SF-G. Because that's what Synomenon was asking about.

I think that, again, you are mistaking the numbers in the review for output and not load-from-wall, while also forgetting the CPU was overclocked in the review. So it wouldn't be at "the very limit" - it would be 77% load which is what you yourself just said was recommended.

ekuest said:
...and the knowledge that the thing could blow if i plug in too many usb peripherals. he could think "oh well metro 2033 only pushed that rig to 358W and im using a lower TDP proc so i will be fine" and then he fires up crysis 3 without remembering to unplug his external hdd and boom dead psu, and probably a motherboard or cpu with it.
Click to expand...

Right, now you're just exaggerating matters. The Bequiet BQT-350W has overcurrent protection. Nothing will blow up, the unit will shut itself down at 400W and has been tested to work fine until then. (OC3D tested it to 420W peak output even.)

ekuest said:
i kept it simple before because i didnt want to have to write all that, but i think its irresponsible to say "stick with that one because the number is smaller" when there's so much more to take into account. it has nothing to do with me distrusting [H] and everything to do with seeing hundreds of computer problems on these forums, and knowing that so many of them resulted from psu issues.
Click to expand...

Those problems are usually because people buy $20 PSUs with no overcurrent protection. We're talking about a quality unit that has been tested to be able to provide above and beyond the rated power.

You can call it irresponsible, but based on the components Synomenon mentioned, there's absolutely no reason to suspect that he will get anywhere near the maximum capability of the unit.

As well as you seeing hundreds of computer problems on the forums, there are also hundreds of posts where someone asks for PSU recommendations and people immediately reply that they need a 750W unit for no apparently logical reason. All you need is a quality unit, and to read the reviews and see that the vdroop is fine at load and that it will shut itself down before it draws too much power or gets too hot. If you don't know what your power consumption is then a kill-a-watt meter is a fine investment which will last you for many systems.
 
I didn't bat an eye when you guys mentioned buying hardware with campaign funds. I think if you are worried from backlash that when finished testing you use it in a promotional way to draw interest in the next fund raiser. Like buy 1 of 100 spots for an ncase and have a chance to randomaly be selected to win a gtx 690.
 
Not sure if it was already mentioned, what usb3.0+audio module you plan to use? One that's already available or one that's custom made for this case?
 
I personally see no reason not to buy current hardware, if they wanted to. I mean they put in all the time and effort, and are going to be selling the M1's for pretty much cost (at least that was the original intent). Unless they plan to turn this into a full-time company, and produce more than they originally planned for, then I see no reason for them not to be rewarded with the toys from testing. After all, these were donations to get the project going. They overachieved with still 20 days to go. There should be enough for a couple revisions, if necessary, and even a GTX 690, or Titan, if they wanted :D

Just out of curiosity, what do you guys plan to do with the extra $$, if there is a lot left over? I'm also curious what donaters and future M1 customers think they should do?

One idea that I have is pooling the leftovers and putting it towards the production run to drive down costs per unit - that is if you guys don't plan to create a full time business from this and get to be rewarded with the left over current gen testing hardware. In no way am I saying you have to do anything... I just think you guys deserve to be rewarded at the end, in some way besides recognition.
 
wahaha360 said:
The panels on M1 are 1.5mm thick b/c Lian Li's machines can't do hole patterns if thicker than 1.5mm, unless we pay for tooling (up to new Porsche expensive).

The challenge is the limitation of existing manufacturing capability. The minute you have to invest in machine tooling, the setup cost just skyrockets, then you would need a large quantity sold for the price to be reasonable.

Unfortunately, this is the reality prohibiting more innovation.

Until SFF becomes main stream, this is a niche product in a niche market, a Niche^2 CASE.
Click to expand...

Have you considered Abee - http://abee.co.jp - as an OEM? Their cases are probably of the highest quality currently available. And by available I mean only in Japan, unless you're willing to unreasonable shipping and taxes. However for a large enough order it might viable. I'll buy an NCASE M1 irrespective of the OEM, but if you could get Abee to do it you'd really have the cream of the crop.
 
Moogles said:
Have you considered Abee - http://abee.co.jp
Click to expand...

A case from Abee is likely to be much much more expensive.

There is an unspoken rule that your product cannot be priced cheaper than those of your manufacturer. If you break this rule, it's unlikely any company will work with you. If Abee is making this, it is highly likely M1 will be $200+ and that price comes at a cost against our goals / values.

Our goals / values for starting NCASE:

wahaha360 said:
We probably could price it higher. In fact, I think it will make some people happy if we did.

Our goal for the M1 is:
- Promote Small Form Factor PC.
- Promote more variety and competition for the mITX market
- Promote more R&D coordination with forum community from companies
- Create a computer case that will push for smaller high-performance components
- Pave the way for others to crowdfund computer components
- Make Necere a real product designer (his 10+ years hobby)

Those goals are best achieved if we make M1 accessible to people, an expensive case is less accessible to people - not everyone has $200 for a case.
Click to expand...
 
Diverge said:
Just out of curiosity, what do you guys plan to do with the extra $$, if there is a lot left over?
Click to expand...

We lost 10% of the funds to Indiegogo, Paypal and bank fees.

Then there is prototyping costs, which starts at $2500 assuming prototype is 95% correct (see Indiegogo Campaign > Fund Allocation for more details).

When we sell the testing components on eBay, we will probably loose 15-30% in the value of components, another 12-15% to eBay and Paypal fees, .

We like to pay some tooling costs to lower the production cost of the M1.

Whatever is left over, ideally, we like to offer it as credit towards backing M1 production run. However, I CANNOT PROMISE because things can go wrong during testing and it's impossible to estimate costs until we know how good the 1st prototype is.
 
wahaha360 said:
We lost 10% of the funds to Indiegogo, Paypal and bank fees.

Then there is prototyping costs, which starts at $2500 assuming prototype is 95% correct (see Indiegogo Campaign > Fund Allocation for more details).

When we sell the testing components on eBay, we will probably loose 15-30% in the value of components, another 12-15% to eBay and Paypal fees, .

We like to pay some tooling costs to lower the production cost of the M1.

Whatever is left over, ideally, we like to offer it as credit towards backing M1 production run. However, I CANNOT PROMISE because things can go wrong during testing and it's impossible to estimate costs until we know how good the 1st prototype is.
Click to expand...

Thanks for replying. I was just curious and totally understand planning for unknowns, it's rare for things to go smoothly. But hopefully they do!
 
Can I ask what the plans are regarding testing watercooling components?

I am looking at http://i.imgur.com/OpdANgA.jpg and wondering how you propose to connect the tubes to the fittings when the radiator is mounted. Also I mentioned earlier in the thread that the Tecnofront Nano is discontinued, but I have yet to see another solution for reservoir placement or T-line. The only option seems to be a radiator with built-in res, such as included with the Swiftech H220. If this turns out to be the most viable solution, it might be worth contacting Gabe @ swiftech to let him know about the NCase project. He's an enthusiast just as much as the rest of us and might be willing to donate an H220 for testing.
 
WiSK said:
Can I ask what the plans are regarding testing watercooling components?

I am looking at http://i.imgur.com/OpdANgA.jpg and wondering how you propose to connect the tubes to the fittings when the radiator is mounted. Also I mentioned earlier in the thread that the Tecnofront Nano is discontinued, but I have yet to see another solution for reservoir placement or T-line. The only option seems to be a radiator with built-in res, such as included with the Swiftech H220. If this turns out to be the most viable solution, it might be worth contacting Gabe @ swiftech to let him know about the NCase project. He's an enthusiast just as much as the rest of us and might be willing to donate an H220 for testing.
Click to expand...

or you choose one of the small bitspower reservoirs, like this or this for example
 
ipuoL said:
or you choose one of the small bitspower reservoirs, like this or this for example
Click to expand...

For my FT03-mini I made a cardboard model of the 40mm one and couldn't fit it inside the case, and that's with a smaller rad that's not mounted above the motherboard. It would have to go on the back of the case.
 
yes, like the one in the renders. Inside is very difficult and as you said i think only possible with an radiator / reservoir combination
 
WiSK said:
Can I ask what the plans are regarding testing watercooling components?

I am looking at http://i.imgur.com/OpdANgA.jpg and wondering how you propose to connect the tubes to the fittings when the radiator is mounted. Also I mentioned earlier in the thread that the Tecnofront Nano is discontinued, but I have yet to see another solution for reservoir placement or T-line. The only option seems to be a radiator with built-in res, such as included with the Swiftech H220. If this turns out to be the most viable solution, it might be worth contacting Gabe @ swiftech to let him know about the NCase project. He's an enthusiast just as much as the rest of us and might be willing to donate an H220 for testing.
Click to expand...

The H220 is definitely something I'd like to test. I suppose it's worth dropping Gabe a line to see if he's game to loan us one.

Fitting a res internally is possible under certain circumstances. For example, if you're watercooling a short card like a GTX 670, this Koolance DDC top and their shortest tube res (30mm) fits in the front:




But for longer cards, you're pretty much limited to using an externally-mounted res or t-line. Here's a Koolance res on the back:



You'll also note when watercooling the video card there's enough room for a DDC pump on the bottom.

There's also this Bitspower 3.5" bay res which could be put in place of the 3.5" HDD on the bottom, except being at the lowest point of the loop it probably won't work well. Maybe if the pump is flipped on its top and the res raised a bit, so the pump intake is below the res. Koolance does something like that with their 5.25" bay integrated pump reservoirs.


I'm probably in the minority, but I think part of the fun of getting watercooling into a tiny system is figuring out how to make everything fit. I've seen watercooling in Shuttle systems, the SG05 and SG07, etc. None of which were ever designed for watercooling. The M1 aims to make it achievable with little or no modding, but it's still going to be a challenge. It's really not possible to make something this small completely watercooling friendly.
 
Synomenon said:
Ah so only the person who contributes $2000 gets a silver version...
Click to expand...
Since that person funded a huge chunk of the project all by himself before the goal was met, this is the least we can do for him or her. We should all be grateful for people like that person, that with or without the disposable income, he or she did finance a project that is not yet even prototyped, so my hat goes off to you who did this.
 
Synomenon said:
Ah so only the person who contributes $2000 gets a silver version...
Click to expand...

So.. what's the problem? The entire concept of the case is a production of black models only to keep it as affordable as possible. If someone was generous enough to put that kind of money in, it's only through a nice gesture that they decide to provide that person with a silver model. They don't have to.
 
XNine said:
So.. what's the problem? The entire concept of the case is a production of black models only to keep it as affordable as possible. If someone was generous enough to put that kind of money in, it's only through a nice gesture that they decide to provide that person with a silver model. They don't have to.
Click to expand...

Agreed. I also don't understand the questions about the donations and what is being done with them. No where in indiegogo or kickstarter does it say you will get an itemized list of what they will spend the money on. It should be understood that the donations are going towards this "effort" and that's all. imho of course
 
RiZnO said:
Agreed. I also don't understand the questions about the donations and what is being done with them. No where in indiegogo or kickstarter does it say you will get an itemized list of what they will spend the money on. It should be understood that the donations are going towards this "effort" and that's all. imho of course
Click to expand...

I agree very much. There is a LOT that goes into the development of a case. I've tried my hand at it, and there are so many facets that you have to think about. Dimensions of each part of the case, dimensions of gear, what you can do to minimize cost yet provide an excellent product, can you watercool it, how many radiators will fit, what kind of GPU will fit, how will you mount drives... it's insane.

Good cases don't come without a ton of work. I don't see why anyone wouldn't want the designers to buy parts with some of the funding. This is essentially a ground-up project that aims to exceed every other mITX chassis on the market. It's not like these guys have a ton of disposable income to throw at this thing.

While I may work for another company, I ALWAYS appreciate great designs no matter where they come from. This case is designed not only as a LAN box, but for watercooling enthusiasts and modders as well. If I weren't a full-time student and I had some extra money (ie, not eating Ramen the next few days), I'd put in money for this project as well.

That's why I'm hoping for the best for this case. It's an amazing design that I think would come from any enthusiast's head. The challenge is getting it to real-world application.

edit: Additionally, designs like this keep other MFG's on their toes and force innovation and competition. It's better for the consumer in the end.
 
Necere said:
I'm probably in the minority, but I think part of the fun of getting watercooling into a tiny system is figuring out how to make everything fit. I've seen watercooling in Shuttle systems, the SG05 and SG07, etc. None of which were ever designed for watercooling. The M1 aims to make it achievable with little or no modding, but it's still going to be a challenge. It's really not possible to make something this small completely watercooling friendly.
Click to expand...

Then I'm in the minority too. I recently built a CPU+GPU loop with the Apogee Drive II in an FT03mini, which you'll know has the inner frame of an SG05. Fun challenge, I learned about annealing acrylic tubing; and in the end had to use a temporary, ghetto, tupperware reservoir to bleed the system.

But although I see how a 240 radiator fits on the M1, I'm still left with the question how to connect the tubes and/or fittings? Your hands will be too big to get down there between motherboard and rad. Or if you are proposing attaching fittings and tubes first, then moving the rad into place, then the tubes are going to kink when they fold back on themselves (and take a lot of space). I just can't see how it's going to work.
 
You must log in or register to reply here.
Back
Top