Performance relevance of old high-performance cases

Stuen4y · 18 Jul 2020 at 10:45

I have been a happy owner of a Fractal Design Arc Midi (R1) since 2013 and it was the best bang for performance and water cooling options case to have 8 years ago. I've updated the fans with 6 Akasa Viper 140mm fans so the airflow of the case is really strong with the fans ramping up based on CPU temperature due to my archaic motherboard. I've also equipped the optional side panel from the R2 and removed the foam at the top (might nuke the one in the front too but it's a bit fiddlier and might be destructive too).
The case industry went through the whole RGB+tempered glass period which actually meant that performance didn't improve dramatically and probably even became worse until several years ago when the mesh fronts started returning again and cases like Phanteks P400A, P500A, Fractal Meshify, CM H500, and Lian Li Lancool II Mesh became kings of the performance charts for noise normalised performance due to their very open fine mesh fronts with no foam air filters impeding the airflow.
I have been trying to get some real performance comparisons between older generation cases, say before 2015 (Fractal Define R4/R5, Arc Midi/R2, the Corsairs and NZXTs of the time) and anything that came after 2018 (the mesh revolution). Any help on determining the performance levels are welcome. I am itchy for new hardware upgrade while waiting for Zen 3 + Ampere/Navi 2 and my wife's PC would happily take the upgrade from it's current £27 NZXT Source 210 case which is horrible in so many ways but has served me proudly since 2012. At the same time, I don't really care for the looks of the case and my current one is practical enough for what I'm doing so even after a full component upgrade I would not mind putting all the hardware in the same case if there is only marginal performance difference with newer chassis options. I'm sure there are tons of people in the same situation as me but couldn't really find any meaningful information about it anywhere. I respect Steve from GamersNexus a lot and their case review methodology is great but they don't have data for old cases and articles like https://www.gamersnexus.net/guides/3604-best-gaming-pc-cases-for-airflow-in-2020-right-now are great if you're buying a case right now without considerations of owning something already. I hate e-waste and don't want to contribute to it unnecessarily (see my ancient CPU for reference

)
Thanks in advance for the participation in the discussion.

Stuen4y · 18 Jul 2020 at 16:49

I'm mostly concerned about the resistance the badly designed foam filter has to the airflow. I was looking at magnetic filters covering the front panel so I can remove the foam keeping dust outside but it's almost better to buy a whole new case. At the same time newer chassis have muche better clearance in the front of the case allowing for unrestricted fresh air. I got more than 5 degrees down when I removed the top exhaust foam of my case so I am wondering how much performance especially for the graphics card is left on the table.

Stuen4y · 18 Jul 2020 at 19:44

I know that Arc Midi was not high end in terms of build quality (it's cheap ugly plastic after all) but at the time cooling performance was quite good with only £200+ cases being ahead and not in a significant way. I strongly believe that the way to achieve quiet operation is via well ventilated case with fewer and slower spining fans rather than noise dampening. Did you observe any significant temperature drops when upgrading from Arc midi to R6?

Stuen4y · 18 Jul 2020 at 20:53

PieEater said:
In all honesty no, I didn't notice any 'significant' changes to temps moving from the Arc Midi, but it had become a bit of a Frankenstiens monster with externally mounted fans and the lower fan cage drilled out in order to acheive the best cooling I could. I would reccomend the R6 more on a compromise between noise & performance and am more than happy with it, I can only hear my fans when gaming or under CPU intensive tasks the rest of the time it is just at ambient noise levels which is great. The R6 is so much better put together you would hardly beleive they are from the same company and in terms of ownership I am much happier with it.

That's good to know. I had that type of a gut feeling. Would be good if there are any other data points abou that type of comparisons.

Stuen4y · 19 Jul 2020 at 16:09

I saw quite the impact of removing the top exhaust foam filter from my Arc Midi , especially on the GPU temps (5+ degrees down with same fan speeds) so would assume the impact on the intake would also be quite dramatic. My PCs front I/O is not used at all as I have powered USB hubs on the desks so that is not a consideration. I don't really care about design either, just expansion airflow, expansion options for storage, and ease of build in that order. I thought I'll build a water cooled setup when I bought my current case 7 years ago but I haven't yet so I guess that's not a priority either.

Stuen4y · 19 Jul 2020 at 18:00

Yeah, Vega 56 is a bit of a monster when it comes to heat so I presume it had to do with trapped heat in a weird way with the air not properly circulating around the GPU. The fans I have installed have very high static pressure and very high airflow so it's not the fans for sure. That's one of the nastiest things with cases, very few people are experts in fluid dynamics so each system is sadly unique.

Stuen4y · 24 Jul 2020 at 07:06

I've decided to do some modding as I've never done anything complex. I'll remove the HDD cage riveted to the bottom and mount my 4 SSDs in the 5.25" bays. I'll also remove the front foam and replace it with some fine mesh on magnetic strips. Might also play with fans position as I might have done turbulence next to the CPU cooler.

Stuen4y · 24 Jul 2020 at 07:18

The new cases have much less airflow obstruction with the death of HDDs and optical drives providing the ability to mount 2.5" SSDs anywhere. This makes the front of the cases empty greatly aiding airflow. That's the main difference in my opinion. I'll report the changes in temperatures after I've emptied the front of my Arc Midi.

Stuen4y · 24 Jul 2020 at 08:25

Yeah 600T was the gold standard, I remember there were few that were better and not absolutely ginormous.

Stuen4y · 28 Aug 2020 at 12:30

Just a small update after I did the mod. I did remove the HDD cage of the Fractal Arc Midi and saw great results. Here are some graphs showing a typical game scenario (in this case Project Cars 2 as I don't need to play it but it offers a repeatable high load scenario) that pushes both the GPU (at 100%) and CPU (at 30-40%). Data was recorder with HWInfo and plotted with PowerBI when it comes to the performance of the case before and after the HDD cage was removed. I had removed all the drives and brackets from the bottom cage prior to this and my 2.5" SSDs sit snugly in the 5.25" bay (which actually resulted in a 2-3 degrees drop by itself but I am not sure if I recorded the full set of sensors when I did that). Both tests were performed with an ambient temperature of 20.5 degrees C given by my cheap room therometer so I'd say +/- 1 degree at least.

With the same voltage for the GPU and CPU set in the BIOS and Radeon Adrenaline (and oddly slightly higher CPU freq but my mobo is not very stable on BCLK) I got almost 6% lower GPU temps, 2% lower CPU temps, lower voltage and power for the GPU (due to the lower temps) and slightly higher GPU clocks. All of that with the added benefit of 13% lower fans for the GPU and 7% for the rest of the system fans (I'm using 6 Akasa Viper 140mm 1800RPM fans for the case and 1 for the CPU cooler and they are all tied to the same PWM header so have synchronised fan speeds). I definitely can hear the fan speed difference but due to the old MB I have, I cannot finely control the fans so I will not be able to exactly reproduce the first test in a controlled environment. I stopped the second test before it hit 1 hour because temps had stabilised and I saw no point of running it any further. Actually GPU temps were perfectly stable throughout the no HDD cage test. The other interesting bit was that after I removed the HDD cage the pressure of the airflow switched from negative to positive (tested with three incense sticks for smoke generation) with very serious airflow coming out from the PCI slots below the GPU which was getting nice fresh air from the front and bottom. This means that the HDD cage was such a great obstruction that the bottom front fan was extremely suffocated and was not working near its pressure and airflow potential.
In summary, the HDD cages at the front definitely make a difference and I recommend to every owner of an old case that still works okay for them to consider removing the 4-8 (6 in my case) rivets with a drill and a 3mm bit, it took me probably about 15-20 minutes with disassembly, drilling, and cleaning. I have purchased a 5.25" to 2.5" bay (which is on the way from China) that should safely hold up to 6 SSDs so that is absolutely fine for me. Once I upgrade and move towards M.2 drives I might remove the 5.25" cage as well but I see less point in doing so due to lack of fan holes there. I will also not be upgrading the case as I think it still has very decent performance and I will not get much more from a newer case apart from build quality and aesthetics. I think from the moment I started tweaking the case airflow I've dropped about 10 degrees of the GPU temps which are definitely the biggest contributor to the heat in the system being over 250W with the CPU definitely below 80W and running cold due to being with Liquid metal between the die and the IHS.