My extensive testing with H4N/NTG-2 with detailed results

markblasco

markblasco

Well-known member
OK, after all of the threads about the H4N and NTG-2, as well as the debate about getting good audio on the Canon 60D, I decided to do some extensive testing to see what I could see (or rather hear what I could hear), to find the best combination. Here is the equipment I was working with:

H4N recorder
NTG-2 mic
Shure SCM268 mixer
Canon 60D
Canon HFS100

The results will be in post#2 for those people who don't care about the process, and just want to know how the story ends.

In order to make sure the test was as repeatable as possible, I took the intro from a podcast I had recorded a while back, which featured both me talking and some intro music. I put this at a level that peaked around -18dB with the vocals. I then created a track of white noise that peaked just below 0dB. I set this to play through one of my studio monitors, and placed the NTG-2 on a mic stand about 3 feet away. This would ensure that the source feeding the test was exactly the same every single time.

Considering that the 60D, HFS100, and H4N all have different level readouts, I calibrated everything by playing the white noise through the speaker and getting that to just below clipping level (for the mixer, cameras, and recorder). This worked well enough to get all of my recordings within 5db of each other, which for this type of test I felt was close enough.

My test also included silence as I stood in the back of the room for 10 seconds or so, in order to hear the noise floor. I am in a room converted to a recording studio in my house, so it is very quiet, and the noise on the recordings was electronic and not noise I could hear in the room. I did have some lights on, so that may have added a little noise (Sometime in the future I will probably repeat the test with all the lights off to see, but for this test that wasn't an important factor, since they were all left on for all of the tests).

I brought all of the audio files into Reaper (for the camera footage I first extracted it to WAV in Premiere). Everything was 24/48k. I took one particular spike in each recording and reduced the levels in all tracks until it was exactly -20dB in every track. The difference in volume adjustments between tracks varied by less than 5dB.

For each camera, I tested with the mic plugged directly in with an XLR to 1/8" adapter, and then the mic running into the Shure mixer and then into the camera through the same XLR adapter. The mixer was running a mic level signal out.

For the H4N, I tested the mic directly into the XLR input with a battery (no phantom power on H4N), no battery (phantom on the H4N) and also through the XLR to 1/8 adapter into the external mic input. I then tested it through the mixer (set to mic level) into both the XLR and the 1/8" inputs (using the XLR to 1/8" adapter), and then tested with the mixer set to Line level out at two different gain settings for the H4N.

With all of these audio tracks, I listened for sound quality, and also was able to measure the noise level. Results in the next post.
 
Results:

Noise levels will be listed in the next post.


Sound quality:

First of all, for most of the recordings, they sounded almost identical. There were a few slight differences, and one major difference.

60D no mixer: This sounded terrible. It was so bad, I'm going to go back and do it again at some point soon, because I can't believe it would be this bad. There was almost no bass, a very nasty high end, and TONS of noise. If I hadn't double checked everything each time, I would think that I made a mistake, but I am 99.9% certain that everything was correct.

60D with mixer: This sounded better, but still worse than all of the other combinations. There was a lack of low end, although not so much that you couldn't fix it later with an EQ. I would use this for a web video and it would probably sound good enough, but it's not ready for anything professional.

HFS100 no mixer: This sounded good, although maybe not quite as good as the H4N (although definitely very very close). I would use this for a professional recording. Slightly more low end than with the mixer (and significantly better than the 60D)

HFS100 with mixer: This also sounded good. Slightly less low end than the mic plugged directly in, but otherwise it sounded almost identical.

All of the H4N recordings sounded about the same, with the exception of whether or not the mixer was used. The mixer seemed to remove some of the low bass. Not a lot (you wouldn't notice unless you were listening for it), but on some decent headphones, you can tell the mixer reduced the low bass slightly. All of the recordings had a little more low end than the HFS100, with slightly more treble (in a good way)

H4N no mixer XLR input battery (rec. level=80): This sounded good.

H4N no mixer 1/8" input battery (rec. level=52): This sounded pretty much identical to the battery XLR input test. I could not hear any discernible difference between the two.

H4N no mixer XLR input phantom power (rec. level=75): This actually sounded slightly worse than with the battery, with significantly more noise.

H4N mixer (mic level) XLR input (rec. level=40): This had a little less bass than the tests without the mixer, and a tiny tiny bit less high treble. You wouldn't notice the bass was lowered unless you listened to both back and forth with decent headphones on.

H4N mixer (mic level) 1/8 input(I forgot to write down the record level): This sounded pretty much identical to the mixer into the XLR input, with a slight bit more treble (although you wouldn't notice unless you were listening for it).

H4N mixer (line level) XLR input (rec. level=10): This sounded the same as the mixer at mic level running into the XLR input.

H4N mixer (line level) XLR input (rec. level=1): This sounded the same as the mixer at mic level running into the XLR input.


Overall, the 60D sounded bad (although usable for low profile gigs when using the mixer), the HFS sounded nearly as good as the H4N, and the H4N sounded the best. The HFS100 was close enough, though, that I wouldn't worry about using that to record the final audio if I couldn't bring along the H4N.

What surprised me is that using the phantom power (set to 48v) on the H4N sounded worse than using a battery on the mic. I had read in several different places that you should do it this way, but the noise level was significantly higher when using phantom power (I will have details below). The XLR and 1/8 inputs both sounded very very similar, so if you wanted to use 3 mics, you could certainly plug the NTG-2 into the 1/8 input, and use the 2 XLR inputs for 2 other mics (and record in 4 channel mode)

Noise levels will be listed in the next post.
 
Noise level results:

This is the area in which most people experience issues. The NTG-2 is known for having a lot of noise when used with the H4N. How you set it up, though, makes a HUGE difference.

All of the noise figures listed below are the noise floor with the audio recording peaking at -20dB. The audio file I was recording had been compressed before playback, so this would realistically line up with a live audio recording peaking around -15dB or so, which is a good real life figure.

60D no mixer: -52.3 (this was the worst result)
60D with mixer: -69 (this was with the gain on the camera turned all of the way down to zero, and then bumped up one)
HFS100 no mixer: -68.2
HFS100 with mixer: -70
H4N no mixer XLR input battery (rec. level=80): -66
H4N no mixer 1/8" input battery (rec. level=52): -68.5
H4N no mixer XLR input phantom power (rec. level=75): -57.3 (second worst)
H4N mixer (mic level) XLR input (rec. level=40): -70
H4N mixer (mic level) 1/8 input(I forgot to write down the record level): -69.4
H4N mixer (line level) XLR input (rec. level=10): -69.6
H4N mixer (line level) XLR input (rec. level=1): -67.6

OK, so the H4N recordings, despite having slightly different numbers, all sounded the same when listening to the hiss, with a few exceptions. Using phantom power caused a lot of hiss, and using the battery into the XLR input had slightly more hiss than the rest of the results. The cleanest result I got without using the mixer was the mic with a battery into the 1/8" input.

The tone of the hiss for most of the H4N recordings, as well as the HFS100 and the 60D with mixer, was all pretty much the same, which leads me to believe that this hiss is coming from either the mic, or it is hiss coming from the speaker. Either way, it is consistent enough in all of the clean recordings that I am confident it is not something caused by the recording device itself.

Overall:

Overall, the best sound quality seemed to come from using either the mic with battery into the 1/8" input on the H4N, or using the mixer into either of the inputs. The HFS100 came very very close, and I wouldn't hesitate to use that instead of the H4N to get audio recording. The 60D...well...let's just say that I will no longer tout the manual gain control as a feature, since the audio is pretty bad. I am a bit saddened by this. I wonder how much the quality would improve if you used a nice mic preamp like the MixPre.

In the future I am thinking of trying a similar test again using my API studio preamps, which are very very quiet and sound fantastic. That should help me to figure out exactly where the noise is coming from. For now, though, this helped to clear up a lot of things for me in terms of how I go about recording audio from the NTG-2 in the future.

Let me know if you have any questions about anything here. I'm not going to redo any of this, since it took a long time to do it in the first place, but if something I wrote wasn't clear, let me know and I will try to fill in any gaps.
 
markblasco said:
Noise level results:
The NTG-2 is known for having a lot of noise when used with the H4N. How you set it up, though, makes a HUGE difference.
Click to expand...

Any thoughts on what mic would perform better with the H4N?
I'm still in the market for both an external recorder and shotgun mic in about that budget if possible to get something workable, and upgrade later if necessarily.
 
I don't have any experience with other shotgun mics at all, so I couldn't say. Using the XLR to 1/8" input though, I found it to have as low a noise floor as with any other combination that I tested, and using the battery yielded OK results as well, so I don't know that you necessarily need to totally rule out this combo. I'm sure someone else here can chime in about something that works better in the same price range.
 
You must log in or register to reply here.
Back
Top