Completed S401: Salvo Studios <8L mITX Steel Unibody Chassis w/ GPU & CPU fans in same direction

[grsychckn]

I see the acrylic panels are a separate ordering process. Does this mean the case comes with the standard panels that we then can swap in and out with the custom acrylic, or does one order override the other?

The steel side panels are included in all sales for the case. I'm going to have 3 pre-order options:

1. Basic case (includes all mounting accessories, steel stands, and steel side panels).
2. Basic case (everything from #1) plus custom acrylic side panels (your color choice).
3. Basic case with custom acrylic side panels (everything from #2) plus removable handle.

There will also be an option to pay an additional fee for your custom logo etched on the solid side panel.

 

[Sparkles]

The steel side panels are included in all sales for the case. I'm going to have 3 pre-order options:

1. Basic case (includes all mounting accessories, steel stands, and steel side panels).
2. Basic case (everything from #1) plus custom acrylic side panels (your color choice).
3. Basic case with custom acrylic side panels (everything from #2) plus removable handle.

There will also be an option to pay an additional fee for your custom logo etched on the solid side panel.

And just to be clear, there will be the option of buying the basic case and handle? Unfortunately, I have no interest in the acrylic sides at this time.

 

[grsychckn]

And just to be clear, there will be the option of buying the basic case and handle? Unfortunately, I have no interest in the acrylic sides at this time.

I have been on the fence with this option because I wasn't sure how many people would want this - but since it doesn't take much to add a new product with this combination I'll go ahead and do so. updated list of pre-order options:

1. Basic case #1 (includes all mounting accessories, steel stands, and steel side panels).
2. Basic case #2 (includes everything form Basic Case #1) plus removable handle.
3. Custom case #1 (everything from Basic Case #1) plus custom acrylic side panels (your color choice).
4. Custom case #2 (everything from Custom Case #1) plus removable handle.

 

[annasoh323]

I have been on the fence with this option because I wasn't sure how many people would want this - but since it doesn't take much to add a new product with this combination I'll go ahead and do so. updated list of pre-order options:

1. Basic case #1 (includes all mounting accessories, steel stands, and steel side panels).
2. Basic case #2 (includes everything form Basic Case #1) plus removable handle.
3. Custom case #1 (everything from Basic Case #1) plus custom acrylic side panels (your color choice).
4. Custom case #2 (everything from Custom Case #1) plus removable handle.

Bit more of a logistical hassle on the ordering/shipping side of things but I think it provides a good level of extra flexibility. Basic case + handle would make the ultimate portable combo since someone who's out and about with the handle might not want to risk scratching their mirrored acrylic. (Side thought: I wonder how a mirrored acrylic vented side would look with RGB? Hmmm...)

You don't happen to have any loading experience with the handle, do you? i.e. we shouldn't exceed a total weight of X amount when carrying with the handle. I suppose a lookup of that screw size should give a rough estimate.

Another thought: How hard is it for you to put multiple logos on the non-vented side? I'd hate to remove your Salvo logo from your product. Would it be possible to request a main image centered but then have the smaller version of your Salvo logo in the corner? My guess is that setup time will be greater but fab time would be about the same?

 

[Sparkles]

Bit more of a logistical hassle on the ordering/shipping side of things but I think it provides a good level of extra flexibility. Basic case + handle would make the ultimate portable combo since someone who's out and about with the handle might not want to risk scratching their mirrored acrylic.

Yeah I guess I could have been more clear on my reasons, but this is exactly right as I'll just need a hardy steel rectangle to pack around. :-)

 

[annasoh323]

Yeah I guess I could have been more clear on my reasons, but this is exactly right as I'll just need a hardy steel rectangle to pack around. :-)

A rare moment of mental clarity for me then .

OTOH, if you were in a situation where you could leave the acrylic panels on in one place for a while, it looks like they come off via thumbscrews such that you could swap easily... So you could have your cake and eat it too. <mental clarity destroyed by thoughts of cake>

 

[Poblopuablo]

There will also be an option to pay an additional fee for your custom logo etched on the solid side panel

Just to clarify, when you say "solid side panel" you mean acrylic right? You're not planning to custom etch the metal pannels are you? (Idk if that would work.... But just wondering to clarify)

 

[grsychckn]

Just to clarify, when you say "solid side panel" you mean acrylic right? You're not planning to custom etch the metal pannels are you? (Idk if that would work.... But just wondering to clarify)

Good question. I did in fact mean the acrylic panel. However, it is possible (though I haven't tried yet) to laser "engrave" the black metal panel as well because the laser would burn off the powder coat to expose the metal underneath. The problem with this is that the metal used to form the case is galvanized sheet metal that's been chemically treated for the paint and probably won't look great after the powder has been removed. That said, I've not attempted it yet because I haven't had the time to try it on a previous prototype. I wanted to etch on the main body though because I wanted to label the holes in an attempt to avoid confusion when installing hardware - so I'll be trying this soon and probably post a picture when complete.

Even if it works well, it would probably be niche unless you like the look of raw galvanized steel. If I was only performing an outline of a shape/text, that might look fine if the lines were thin enough. Long story longer: I think I need to test it with different thicknesses/shapes to see what it really looks like before I offer it as a service.

 

[annasoh323]

Folks that have the S400: have you experimented with open vs blower coolers in your version of this case? On the one hand, scattered exhaust from an open video card would ostensibly raise CPU temps. On the other, the case seems open enough where it mightn't make an appreciable impact. Any first hand observations? I'm locked in with what I have now (leaf blower 1080) but like most folks, always considering the next upgrade.

 

[papayo]

Folks that have the S400: have you experimented with open vs blower coolers in your version of this case? On the one hand, scattered exhaust from an open video card would ostensibly raise CPU temps. On the other, the case seems open enough where it mightn't make an appreciable impact. Any first hand observations? I'm locked in with what I have now (leaf blower 1080) but like most folks, always considering the next upgrade.

Using the 3dMark Fire Strike Stress Test:
- GTX 970 blower style card: gpu failed the test with a 91% frame rate stability score; GPU likely throttled as it "stabilized" at 80c; CPU stabilized around 64-68c
- GTX 1070ti "open" card: GPU passed the test with a 98.9% frame rate stability score; GPU temps averaged 72c; CPU stabilized around 72-76c
- System temp under stress seems to be at least +4c with the open card (52c vs 56c).

These results are with no case fans in the system.

 

[annasoh323]

Using the 3dMark Fire Strike Stress Test:
- GTX 970 blower style card: gpu failed the test with a 91% frame rate stability score; GPU likely throttled as it "stabilized" at 80c; CPU stabilized around 64-68c
- GTX 1070ti "open" card: GPU passed the test with a 98.9% frame rate stability score; GPU temps averaged 72c; CPU stabilized around 72-76c
- System temp under stress seems to be at least +4c with the open card (52c vs 56c).

These results are with no case fans in the system.

Interesting, I've not tried the stress test; thanks for sharing. I was thinking about it and I'm not sure how I would answer my own question. How does one establish a baseline for such a test? Weighted average based on performance and noise? Some kind of performance/framerate cap? This is why I didn't get very far in the district science fair back in middle school (granted, I didn't get very far in any field back in middle school...)

Thinking about this is leading me to think I should do some kind of migration comparison from the RVZ02 to the S401 (letters! numbers! excitement!) but the S401 doesn't support my existing fan... so no apples to apples comparison here. Yet I'd hate taking apart my RVZ02 build, testing it, ripping it apart, rebuilding in the S401, and testing again... I've not got the YouTube sponsorship $$$ for that. Sorry, science, beaten by life again...

 

[papayo]

Interesting, I've not tried the stress test; thanks for sharing. I was thinking about it and I'm not sure how I would answer my own question. How does one establish a baseline for such a test? Weighted average based on performance and noise? Some kind of performance/framerate cap? This is why I didn't get very far in the district science fair back in middle school (granted, I didn't get very far in any field back in middle school...)

Thinking about this is leading me to think I should do some kind of migration comparison from the RVZ02 to the S401 (letters! numbers! excitement!) but the S401 doesn't support my existing fan... so no apples to apples comparison here. Yet I'd hate taking apart my RVZ02 build, testing it, ripping it apart, rebuilding in the S401, and testing again... I've not got the YouTube sponsorship $$$ for that. Sorry, science, beaten by life again...

Adding noise to the equation definitely complicates the issue considerably. I've actually tried three cards in the S400: the 970 oem blower, an msi 1060 itx, and the EVGA 1070 ti SC. The performance bit is easy to see, but throw in heat and noise and try to distill that into some comprehensive score... well then opinions start to get in the way since everyone's use case is slightly different.

For me:

Once I discovered that the 970 blower was likely throttling, it had to go. Why keep a 970 that couldn't be a full 970.

Personally I think the 1060 itx was the best fit for my system: barely audible at load, compact enough to try to mount additional case cooling, and powerful enough for my purposes at 1080p with zero throttling issues.

The 1070 ti is exactly what you would expect: loud, hot, and more powerful. Once I saw that it didn't burn down itself, my case, or house... It was hard to go back lol. But the acoustics are the worst of the bunch by far. I can hear it from across the room. In another life that would have been a deal breaker, but right now I just feel like overstuffing an itx case

I'm curious if anyone has had better luck with a high-end blower style card.

 

[annasoh323]

Adding noise to the equation definitely complicates the issue considerably. I've actually tried three cards in the S400: the 970 oem blower, an msi 1060 itx, and the EVGA 1070 ti SC. The performance bit is easy to see, but throw in heat and noise and try to distill that into some comprehensive score... well then opinions start to get in the way since everyone's use case is slightly different.

For me:

Once I discovered that the 970 blower was likely throttling, it had to go. Why keep a 970 that couldn't be a full 970.

Personally I think the 1060 itx was the best fit for my system: barely audible at load, compact enough to try to mount additional case cooling, and powerful enough for my purposes at 1080p with zero throttling issues.

The 1070 ti is exactly what you would expect: loud, hot, and more powerful. Once I saw that it didn't burn down itself, my case, or house... It was hard to go back lol. But the acoustics are the worst of the bunch by far. I can hear it from across the room. In another life that would have been a deal breaker, but right now I just feel like overstuffing an itx case

I'm curious if anyone has had better luck with a high-end blower style card.

I was just speculating on the various aspects commonly quoted as an advantage/disadvantage of the two GPU designs and other factors. Somehow, the right way (I think) would be to equalize the heat output of the cards to the heat sinks and let the rest be variables. No idea how to practically set that condition though. Science isn't something that's meant to be done on this level of sleep. On to anecdotal observations!

My only two high-end graphics cards have been blower cards: an MSI 1070 and an EVGA 1080. I'm sure that they both "throttle" with respect to their open air, hybrid, and custom water-blocked brethren and I fully believe them to be louder. However, I've been happy with their performances with my 1440p 144Hz monitor with the games I've played. I've never really maxed out the monitor but I've also not tweaked settings to the max, instead setting things to a moderately high preset and letting the chips fall where they may. I also let my headphones take care of the jet engine sound <insert nervous chuckling>. The 1080 is the one that'll be migrating along with its i5-7600K playmate. I'm treating the 60mm case fans as serendipitous airflow rather than the secret sauce to better temps right now since I'm not sure how much better it'll make things. Maybe @grsychckn has some data on how much of a benefit they might give; otherwise science might strike again...

 

[grsychckn]

I was just speculating on the various aspects commonly quoted as an advantage/disadvantage of the two GPU designs and other factors. Somehow, the right way (I think) would be to equalize the heat output of the cards to the heat sinks and let the rest be variables. No idea how to practically set that condition though. Science isn't something that's meant to be done on this level of sleep. On to anecdotal observations!

My only two high-end graphics cards have been blower cards: an MSI 1070 and an EVGA 1080. I'm sure that they both "throttle" with respect to their open air, hybrid, and custom water-blocked brethren and I fully believe them to be louder. However, I've been happy with their performances with my 1440p 144Hz monitor with the games I've played. I've never really maxed out the monitor but I've also not tweaked settings to the max, instead setting things to a moderately high preset and letting the chips fall where they may. I also let my headphones take care of the jet engine sound <insert nervous chuckling>. The 1080 is the one that'll be migrating along with its i5-7600K playmate. I'm treating the 60mm case fans as serendipitous airflow rather than the secret sauce to better temps right now since I'm not sure how much better it'll make things. Maybe @grsychckn has some data on how much of a benefit they might give; otherwise science might strike again...

Well, I was without internet most of the weekend due to an "upgrade" I took for Comcast service. They disabled my modem because they shipped me a new one and I had to wait for that to arrive before I got the opportunity to reconfigure my LAN because I now HAVE to use their modem (because of voice service that is forced down my throat).

So I've been following your back and forth and have decided to run some tests of my own to see what difference the 60mm fans make. I will tell you this, the Delta 60mm (15mm thick) PWM fans I purchased can move some air. They're fairly quiet but when they need to they kick up the RPMs and take off - they are no joke. I scrapped together a system to use for testing and since 3dmark has been mentioned, I figured I'd use that as a baseline for my tests. I'm going to record the values listed below in each test:

1. Max CPU temp/frequency (CPU0)
2. Max GPU temp/frequency
3. System temp

And I'll be running the 3dmark firestrike stress test using the following configurations (all S401 case):

1. Ryzen 7 1800X w/ 8GB RAM and a WX9100 (Vega-based pro card, blower style) w/ 60mm case fans
2. Ryzen 7 1800X w/ 8GB RAM and a WX9100 (Vega-based pro card, blower style) w/o 60mm case fans
3. Ryzen 7 1800X w/ 8GB RAM and a MSI RX580 Gaming X (open air fans) w/ 60mm case fans
4. Ryzen 7 1800X w/ 8GB RAM and a MSI RX580 Gaming X (open air fans) w/o 60mm case fans

Unfortunately, this is going to take me several days to collect data and then I have to formulate it somehow to present. In an ideal world where I have all the time and sleep I want, I'd also test with the acrylic side panels, horizontal position, as well as contrast it with the S400. I think of those, I'd prioritize the comparison to the S400 so I'll try to make that happen first if I have the time.

Having the S400 and the S401 side-by-side, I don't see how the S401 is worse than the S400. The hex pattern provides more vent area, the material on the S401 is thinner (requiring less effort to pull fresh air in), and there is more volume between the GPU and the motherboard to dissipate the heat.

 

[annasoh323]

Well, I was without internet most of the weekend due to an "upgrade" I took for Comcast service. They disabled my modem because they shipped me a new one and I had to wait for that to arrive before I got the opportunity to reconfigure my LAN because I now HAVE to use their modem (because of voice service that is forced down my throat).

So I've been following your back and forth and have decided to run some tests of my own to see what difference the 60mm fans make. I will tell you this, the Delta 60mm (15mm thick) PWM fans I purchased can move some air. They're fairly quiet but when they need to they kick up the RPMs and take off - they are no joke. I scrapped together a system to use for testing and since 3dmark has been mentioned, I figured I'd use that as a baseline for my tests. I'm going to record the values listed below in each test:

1. Max CPU temp/frequency (CPU0)
2. Max GPU temp/frequency
3. System temp

And I'll be running the 3dmark firestrike stress test using the following configurations (all S401 case):

1. Ryzen 7 1800X w/ 8GB RAM and a WX9100 (Vega-based pro card, blower style) w/ 60mm case fans
2. Ryzen 7 1800X w/ 8GB RAM and a WX9100 (Vega-based pro card, blower style) w/o 60mm case fans
3. Ryzen 7 1800X w/ 8GB RAM and a MSI RX580 Gaming X (open air fans) w/ 60mm case fans
4. Ryzen 7 1800X w/ 8GB RAM and a MSI RX580 Gaming X (open air fans) w/o 60mm case fans

Unfortunately, this is going to take me several days to collect data and then I have to formulate it somehow to present. In an ideal world where I have all the time and sleep I want, I'd also test with the acrylic side panels, horizontal position, as well as contrast it with the S400. I think of those, I'd prioritize the comparison to the S400 so I'll try to make that happen first if I have the time.

Having the S400 and the S401 side-by-side, I don't see how the S401 is worse than the S400. The hex pattern provides more vent area, the material on the S401 is thinner (requiring less effort to pull fresh air in), and there is more volume between the GPU and the motherboard to dissipate the heat.

Whoa! That's amazing. If you're running benchmarks already, you should stake a claim over in the Performance-per-Liter thread. I wonder how much of a threat you'd be over there.

I know that's a ton of work ahead of you, but if I may pose a few clarifying questions:
1) Will you be using the Cryorig C7 Cu as CPU cooler?
2) Performance-wise, where does the WX 9100 approximately fall compared to the consumer lineup? Quick Google search and it looks like it may output heat and gaming-ish performance similar to a Vega 64. Is that close?
3) Are you still using the small Noctua by the video card vent? Sidebar: Here's my assumption about how the fans are connected: CPU fan on a CPU fan header, the two Deltas on a splitter to a system fan header, and the Noctua connected directly to PSU (via noise reducer?) That is, unless you're one of the lucky folks that has a mITX mobo with multiple fan connections.

I'm sorry to cause you more work. Please pass my sincerest apologies to your wife and kids as well.

 

[grsychckn]

Whoa! That's amazing. If you're running benchmarks already, you should stake a claim over in the Performance-per-Liter thread. I wonder how much of a threat you'd be over there.

I know that's a ton of work ahead of you, but if I may pose a few clarifying questions:
1) Will you be using the Cryorig C7 Cu as CPU cooler?
2) Performance-wise, where does the WX 9100 approximately fall compared to the consumer lineup? Quick Google search and it looks like it may output heat and gaming-ish performance similar to a Vega 64. Is that close?
3) Are you still using the small Noctua by the video card vent? Sidebar: Here's my assumption about how the fans are connected: CPU fan on a CPU fan header, the two Deltas on a splitter to a system fan header, and the Noctua connected directly to PSU (via noise reducer?) That is, unless you're one of the lucky folks that has a mITX mobo with multiple fan connections.

I'm sorry to cause you more work. Please pass my sincerest apologies to your wife and kids as well.

1. No, the standard C7 was installed and I'm going to leave it on - besides, results vary so much with the C7 Cu that I think it best to eliminate one more variable.
2. WX9100 is essentially Vega FE slightly lower clocked. The base clock is 15% less than the FE and the boost is less than 10% less than the FE.
3. I am using the Noctua but my motherboard supports 2 chassis fans so I have the Delta (15mm) fans connected to one header, and the Noctual (25mm) connected to the other.

 

[Poblopuablo]

Is it possible to make the side pannel extrude a little to accommodate a slightly larger CPU cooler? (Like stand offs or something)

Thinking about the upcoming raijintek Pallas micro with a 55mm CPU cooler height.

 

[annasoh323]

Is it possible to make the side pannel extrude a little to accommodate a slightly larger CPU cooler? (Like stand offs or something)

Thinking about the upcoming raijintek Pallas micro with a 55mm CPU cooler height.

I've thought about this. Since the side panels are screwed on, I thought it ought to be possible to cut out some simple frames with holes drilled and then obtain some longer thumb screws. I figured the acrylic would be light enough such that i wouldn't have to worry about tipping as long as I didn't go too far. It would make things look fairly uneven I would think... but I was thinking of maybe experimenting in that direction. Of course, all of this heresy is not sponsored by @grsychckn or Salvo Studios (I know, one and the same...).

For the time being, though, since my fabrication skills are nothing much to write home about, I was probably going to stick with my short cooler and two-slot card until I embark on my next big upgrade. Maybe then I can move onto using the Big Shuriken 2 and a triple slot card

 

[Poblopuablo]

I've thought about this. Since the side panels are screwed on, I thought it ought to be possible to cut out some simple frames with holes drilled and then obtain some longer thumb screws. I figured the acrylic would be light enough such that i wouldn't have to worry about tipping as long as I didn't go too far. It would make things look fairly uneven I would think... but I was thinking of maybe experimenting in that direction. Of course, all of this heresy is not sponsored by @grsychckn or Salvo Studios (I know, one and the same...).

For the time being, though, since my fabrication skills are nothing much to write home about, I was probably going to stick with my short cooler and two-slot card until I embark on my next big upgrade. Maybe then I can move onto using the Big Shuriken 2 and a triple slot card

Depending on if you want to have it vertical or not.

I'd be fine with a 1/4"(~4mm) gain.

 

[grsychckn]

I've thought about this. Since the side panels are screwed on, I thought it ought to be possible to cut out some simple frames with holes drilled and then obtain some longer thumb screws. I figured the acrylic would be light enough such that i wouldn't have to worry about tipping as long as I didn't go too far. It would make things look fairly uneven I would think... but I was thinking of maybe experimenting in that direction. Of course, all of this heresy is not sponsored by @grsychckn or Salvo Studios (I know, one and the same...).

For the time being, though, since my fabrication skills are nothing much to write home about, I was probably going to stick with my short cooler and two-slot card until I embark on my next big upgrade. Maybe then I can move onto using the Big Shuriken 2 and a triple slot card

Because the side panel fasteners are #6-32 nuts, you can use standard standoffs (#6-32) to insert space between the chassis and the side panels. I would offer a word of warning though, make sure your standoffs are #6-32 on both the male and female ends - some #6-32 standoffs have convert from the male #6-32 to the female #4-40 (specifically for motherboard mounts).

Depending on if you want to have it vertical or not.

I'd be fine with a 1/4"(~4mm) gain.

If horizontal, the same modification could be done (standoffs inserted between the chassis and side panels). In the horizontal orientation, you could even do this on both sides (in this case the top and bottom) but I'm not sure why one would.