We have a brand new test bench that we assembled for the 2013-2014 period! Having moved away from our trusty i7-930 and GTX 580, the new bench includes the below components:
|GN Test Bench 2013||Name||Courtesy Of||Cost|
|Video Card||XFX Ghost 7850||GamersNexus||~$160|
|CPU||Intel i5-3570k CPU||GamersNexus||~$220|
|Memory||16GB Kingston HyperX Genesis 10th Anniv. @ 2400MHz||Kingston Tech.||~$117|
|Motherboard||MSI Z77A-GD65 OC Board||GamersNexus||~$160|
|Power Supply||NZXT HALE90 V2||NZXT||Pending|
|SSD||Kingston 240GB HyperX 3K SSD||Kingston Tech.||~$205|
|Optical Drive||ASUS Optical Drive||GamersNexus||~$20|
|Case||NZXT Phantom 820||NZXT||~$250|
|CPU Cooler||Thermaltake Frio Advanced Cooler||Thermaltake||~$60|
All of our testing is conducted in a temperature-controlled environment. Ambient is between 21C and 22C for all case airflow tests. The graphs measure temperature in Delta T over Ambient (C) - so the ambient temperature is subtracted from the component temperature.
Each test is initiated with a cold boot, where the system will sit idle for 15 minutes and collect thermal data. We use CPUID's HWMonitorPro for thermal logging and tracking. After this idle time, the system will launch a Prime95 instance running four torture threads on Large FFTs for maximum heat generation and power utilization. This is run for 15 minutes, throughout which the logging utility will collect the data we used in the below charts. Another 15 minute cool-down time (idle) is allowed after the Prime95 instance is completed, at which time an instance of FurMark launches and tortures the GPU with its 15-minute burn-in test (1080p). A final round of idle time is allowed to ensure data consistency.
All CPU thermal results are computed using Core 1 (not Core 0) due to its higher thermal reliability on our IB chip.
NZXT H630 Optimal Fan Configuration Benchmark
We try to test multiple aftermarket fan configurations in these cases when possible. When an enclosure states that it has room for "7 extra fans," or something similar, we expect that a lot of potential purchasers would be interested in which additional fan position nets the most gain.
For this test, we used NZXT's FZ-200 fans for the top/front positions and an NZXT 140mm fan for the bottom position. Here's what we tested:
- Stock configuration (compared against other cases below). 1x200mm front intake, 1x140mm rear exhaust.
- 2x200mm front intake, 1x140mm rear exhaust.
- 1x200mm front intake, 1x140mm rear exhaust, 1x200mm top-rear exhaust.
- 2x200mm front intake, 1x140mm rear exhaust, 1x200mm top-rear exhaust.
For reasons explained below, we also tested all of these configurations without the top and front panels.
I think I'll break the ice with the charts:
Look closely at that. There's almost no advantage in either GPU or CPU cooling to install extra fans -- and that doesn't make any sense. All the results are within our 5% scientific margin of error, so for all intents and purposes, the stock config was effectively equivalent to the aftermarket configs. But why?
Ambient temperature is always controlled in our tests, so that was out. The fans were all manually set to 100% speed via BIOS, so auto-control was out. The CPU thermals tested fine in our recent SilverStone RV02 review, and we haven't remounted the cooler since that test; CPU voltage and OC settings were perfect, and when I ran regression tests of the system (without removing the cooler) in the Phantom 820 and RV02, I saw nearly identical results to what's on our bench... so test error as it pertains to CPU mounting and thermalpaste were both ruled out.
I reran all the tests. The results were the same. After looking back through some of my Phantom 630 review notes and benchmarking data, I saw that the 630 had a 6C improvement to thermals when I'd mounted additional 200mm fans in the same positions. This is noteworthy because, as stated sufficiently, the H630 uses the same chassis -- if the Phantom 630 cooled significantly better than the H630, and both use the same chassis, then that leaves us with paneling issues.
After talking with NZXT's Rob Teller (Case Accessory Product Manager), who helped rule-out test error and component issues, I reran all the tests a third time - this time without the top or front panels. Obviously the thermals were cooler across the board, but the differential between stock and aftermarket configurations was still so insignificant that it was within error. Just like the paneled tests.
The only possible factor I've overlooked is the aforementioned FZ-200 fans. We haven't used them on the bench before, and being that NZXT is rolling these out to replace their old 200mm fans, it made sense to give them a shot on the H630. I didn't use these on the P630, so it's possible that the heart of the issue is here... but if I'm honest, I sincerely doubt that. A fan is a fan. And at 100+ CFM @ 100% RPM, these should be producing results.
Because I've ruled-out test error (by testing the system within the Phantom 630, 820, and RV02 -- which all showed an improvement when extra fans were added), I'm inclined to say that there's something odd going on with the airstream inside of the case. The fans are still a possibility, though I consider that unlikely.
It's an actual mystery, and due to deadlines (I've already spent 3 days too long trying to figure this out), I won't be able to do further testing until the next 2 weeks. If other reviewers (or NZXT testers) reading this would like to help out, here's what I want to do -- feel free to test this yourselves:
Remove the top and front panels from the P630. Run the stock configuration with the original 200mm fans (before the new ones were issued). Add 1x200mm exhaust to the top-rear, 1x140mm in the bottom (intake).
Replicate this test on the H630, compare the differentials.
I originally debated whether I'd even write this entire section -- it doesn't exactly show differences, and that's what you all want to see. It also doesn't seem to coincide with reality or physics, which makes it seem like either the test or the product is bad, and I'm not sure I think that either of those is true. I think it's important to report findings, though, even if they're not showing the expected results (cooler performance) and seem to be conflicting with what experience dictates. As GN integrates more of the scientific method in its testing, hopefully producing objective data for analysis, it's crucial to share what we learn.
Let's move on to the actual case v. case benchmarks. The more immediately useful ones.
NZXT H630 vs. Phantom 630, 820, RV02, HAF X, etc.
I'll just jump right into it:
As shown in the CPU cooling chart above, we see that the stock configuration of the H630 isn't exactly impressive. It hovers around the same level of performance as the 630 and 820 (just below them), as expected, and doesn't really produce anywhere close to the mind-blowing results of the RV02.
Similar findings here. The H630 clocks in at the bottom of our bench, somewhat disappointingly. The idle temps were actually pretty darn good, but at 53.1C Delta T for load testing, it's certainly a far cry from the likes of the HAF X or Armor Evo.
Problems With the H630 & Final Thoughts
Now to voice some opinions. The H630 has particularly solid build quality, even when compared against other NZXT cases we've tested; its thick, padded steel paneling is resilient and sturdy, it has excellent ease-of-installation options, and has good radiator and fan add-on options.
The only point I was decidedly unhappy with build quality was the oddly difficult front panel, though we've been told that was a pre-production defect.
As for actual silence and cooling, well, there's a lot to be desired. A whole lot. There's no disillusionment by any experienced builder that "silence" often equates a detriment to performance. You can't have high-performance and total silence, so that's not news to anyone; you don't buy an H630 or one of Fractal's Silent Series cases and expect SilverStone performance -- that's just not how it works.
That said, there should be a trade-off: If I'm cutting fan count, RPMs, and increasing fan size, I should be receiving a very low dBA emission in return. The thing with the H630, though, is that it doesn't cut RPMs (no integrated fan control and a high-RPM 140mm stock fan), and despite all the foam-adorned paneling, it isn't all that silent. It's a step in the right direction, but it's not there yet; however, as a user, you can take it to that next level. Adding a fan controller would be a good first step.
At the end of the day, the H630 is neither a performance case nor a silent case. It's somewhere in-between, running quieter than the P630, for instance, but still louder than Nanoxia's Deep Silence series and Fractal's Silent Series cases (Define XL, Define R4). Performance is disappointing (and confusing) across the board, performing just below the P630 and P820.
The quality is good and its intentions are there, but it just doesn't perform. My recommendation? Weigh your needs. Do you demand true silence? Look into Nanoxia, Fractal, and aftermarket solutions (like quieter fans in general). Need performance? The RV02 is the winner, and it's not too loud, either.
- Steve "Lelldorianx" Burke.