HM100 - very hissy pre-amps???
 

Got my HM100 on Friday and so far I am very pleased with it - except for one thing. The mic pre-amps on my camera appear to be very hissy. The sound itself is pretty ok assuming that my AKG will be superior to the shotgun that came with it, but I can clearly hear the hiss basically at all audio level settings except when turned to 0. And it´s the same whether I just monitor the audio in the camera or after I imported it to my NLE. Coming from a Canon HV20 and a XH-A1 I´d say the pre-amp is just as bad as the ones in the HV20 - unless my camera is defective. What is your experience? Any problems so far? If I have time later on I will try to post an example...

I thought mine was really noisy but then I saw that I was hearing the noise coming from the automatic gain control on channel 2 if there is nothing plugged in. It tries to turn it all the way up. Changing channel 2's input to channel 1 or setting it to Line and 0 made the noise on that go away.

After that at normal levels I didn't hear very much noise at all. There is a little hiss but I wouldn't say it was "very hissy."

Glad to hear that, Elvis. Please ignore one of my follow-up questions on the other thread re: your experience with 'hissy audio'.

Not having a camera in hand yet, I'm not sure I understand exactly the button and setting options you are referring to, but I'm sure it will be clearer once I take delivery this week.

I tried some higher end studio mics and they were very quiet so I think everything is alright.

So here is an example with the mic that comes with the camera. The audio level is on manual 5-6, I am about one foot away from the mic. The second channel is on manual too and dialed down to 0. The hiss/noise is pretty bad, if you ask me. Is that how your cams sound too or did I get a faulty one?
Thanks,
Matthias

So - do I expect too much or what? Really doesn´t sound good to my ears but maybe I´m doing something wrong? It would be great to hear some comparison from another cam so that I can decide whether to send it back or not...

Hi Matthias.

I've gotten very interested in the HM100 after seeing footage by Phil Bloom and Matt San.

But before I start losing interest in the camera after reading of your audio difficulties, I was just wondering if you'd followed Elvis's advice exactly.

The bit about "the noise coming from the automatic gain control on channel 2 if there is nothing plugged in".

Presumably, if you have a mic plugged in to channel 2 then all should be well.

If not, Elvis's advice was to change "channel 2's input to channel 1 or setting it to Line and 0".

Elvis said that if you dial it down to 0, then you also have to set it to "Line". From a photo of the HM100 audio controls which I found online, the switch for Line is under the label "AUDIO INPUT". Yet you've stated that the "second channel is on manual too and dialed down to 0". According to the photo I'm looking at, the "Manual" setting is part of a different switch labelled "AUDIO SELECT".

Could you just re-check what you've done and report back.

It's important for me to know if the HM100 really does have bad audio hiss.

Thanks.

Audio Test
 

There was more hiss than I thought when I initially tested it but turning off Channel 2's automatic leveling does cut the audio quite a bit. The only thing I did to the audio was normalize it (bring it up to zero where it was mostly at -12 dB)

http://files.me.com/elvis/6lf6fh.mp3

The nicer mics that required less gain and were nearer to me were much quieter so a nicer shotgun would probably help there and something on a boom pole.

Update: I checked and it does make a big difference if the channel is on Mic vs Line but not Mic vs Mic 48V. That makes sense.

Hi David,
I checked all combinations I could think of with no luck. The hiss is almost gone if both channels are on line-in, which is logical since it turns off the internal pre-amp and the only hiss I would be hearing is from the headphone pre-amp. I am not sure what Elvis is talking about though since his scenario would mean that there is some kind of bleeding between the channels going on. In my example you hear just one channel, I deleted the other one in my NLE. And as I stated, both channels were on manual, the one I didn´t use (channel 2) was dialed down to 0. It sounded the same when it was on line-in (which should make no difference unless the would be some kind of bleeding...).
I have a hard time believing that JVC would consider what I am getting "professional sound". So I hope that I just have a defective unit and will see if I can send it back to B&H (unfortunately I´ve already thrown out the box...).

There was more noise than I originally thought. Doing my test in the above post showed the noise I originally noticed but when you take it away there is still plenty of noise there. I will just work around it. I will try a better shotgun and see how that affects the amount of needed gain.

Elvis, I checked you examples and there are some inconsistencies that I don´t understand. I suspect that you have the "CH2 Input" set to "Input1" which would mean that you are mixing the channels? But anyway, could you please do me a hugh favor and post the result of a simple test?
Unplug any mic, just have the handle and the mic cable connected to the cam. Put the "CH2 Input" to "Input2". Put "Audio Input" to "Mic" for both inputs. Put both channels on "Manual" and dial both channels up to 6 and record 30 secs or so. And than dial both channels up to 10 and record 30 seconds. I would be extremly interested to compare that to my cam. Here is what my cam sounds like if I dial both channels up to 10 - the noise floor is somewhere around -42db:

But that´s my point. From my experience you should be able to turn the gain up to at least 5 or 6 before you get any noticeable hiss. Otherwise it just means that the pre-amps are crap. Yes, you can work around it just like I did with my HV20. Been there, done that. But I don´t want to since I spent a lot of money for a camera that promises to be suitable for pro work.

Channel 2 was set on Input 2 for my test.

Here is the results of what you wanted.

http://files.me.com/elvis/fqkq9g.mp3

Wow, that´s a big difference: Your cam dialed all the way up to 10 checks out at around -50db... Seems like there is something wrong with my cam... Should not make a difference, but are you using .mov files or .mp4?
Thanks a bunch!

.mov

For your test I took the QuickTime file and dragged it into Compressor and chose mp3 320 kbits so it should be as close as it can be and still be an mp3.

I will put up the camera file to make sure there is no misunderstanding. .mov or .mp4?

So can someone please sum up the final solution, as I am a bit lost in the maze of these posts?
- Is there a hiss in audio?
-How can you eliminate it?

Every pre-amp produces some hiss. The question is how much. My camera is very hissy, apparently much more so than the one Elvis has. I am going to send mine back for a replacement that hopefully will perform better. Otherwise I will have to go for a different model because it is basically unusable. But I am hopeful since the hiss levels Elvis gets with his cam sound pretty ok and manageable.

I just tried the mic in out auto/manual thing

and basically if the manual control is set to:

line ZERO hiss
Mic +48 auto gain or manual control above zero - minor hiss
mic+48 manual control zero - no hiss

so the mic preamps do create a bit of hiss on mine - but not distracting amounts

I spoke to a friend of my, who for years worked in a recording studio. He was wondering about the microphone on the camera.

And he is right.

Do the same test, but short the mic inputs to ground (make sure phantom power is off first!), turn the mic gain all the way up, turn any other level controls that affect the gain in the camera before the sound is recorded all the way up.

The noise levels in the two recorded files should match between two cameras of the same make and model within at least 20% or so. If not, there is a problem with one of them..

-Mike

The shotgun mic that comes with the camera is super cheap and probably should be replaced. If you want a cheap $3500 camera then it should be fine. If you want a pro quality small camera then you should get a better mic.

There is some hiss but it isn't horrible. You can use some eq to bring down the highs where most of the noise is or get a more sensitive mic so you need less camera gain. You could also get a portable recorder and plug your mics into that.

One would hope that if you spend $3500 for a camera it would have a microphone for $3500 camera, particularly with uncompressed sound. And that was my point a few months ago, that charging customers for worthless microphone is a bad business.

That would pretty much defeat the purpose of the small size of the camera.

Reference Level
 

I replied to another post and realized that if you set the audio reference level to -12 dB instead of -20 db then you get 6 dB of gain without adding hiss. If you have something really quiet this is where you should turn it up first then dial up your analog gain on the handle.

So the consensus on the typical HM100 audio is ...
 

... just plain and very unfortunately Hissy?

Only as much hiss as anything else below $5,000?

Enough hiss to make the uncompressed LPCM a bit too polluted for true professional audio use?

*** The HM100, aside perhaps from a few lemons, has perfectly acceptable audio for professional field use WITHOUT add-on devices aside from mics??

Please describe a bit so that those of us poor souls in the dark that are considering buying one can get a sense of the problem -- or is it a non-problem?

This might be my first camcorder purchase EVER, and I'd like to make it a good one :)

THANKS!

~ Shaun

I think this isn't a problem. Setting the audio level manually keeps the auto gain from turning the level up too much. This is normal for a video camera. I have used it quite a bit and don't have any issue with it.


When you get the camera set the audio reference level to -12 for normal stuff and -20 for louder stuff when you need more headroom to make sure you don't clip.

Set your audio level manually which is what you should do. If you have time to manually set your shutter/aperture and gain then you have time to set your audio level.

You should also get a better mic if audio is important to you. This is also normal for a video camera. For $200 to $300 there are plenty of mics that will really give you a big improvement.

So for shooting with AGC ...
 

So if I am in a run'n'gun situation where I'm shooting everything full auto for a while, and it's in a quiet but rushed situation, the AGC on the HM100 won't be any worse than AGC and full auto on say, a Canon XH A1 or a current Sony prosumer HD cam? ... or what would be a well-behaved prosumer camera that behaves well in full-auto run 'n' gun style when you really need it to, this HM100 or ?

Thanks -- I'm sure looking forward to having two real built-in XLRs, phanton power, and adequate manual level controls and meters. The last time I shot video for more than a bit, it was on 1990ish AG-1xx series cheap industrial-consumer S-VHS Panasonics ... We just had our professor's ME60 plugged into the mono mini Mic jack with a cord that had an attenuator or ? on it.
I plan to get a Rode Rode NTG-3 Shotgun or a Sennheiser ME66/K6 and other modules, probably the latter.

I think it will be about the same. You could have channel one set on auto and channel 2 set to manual 2 just in case the auto goes too loud or the source gets too loud for the auto to handle. It is just a way to have a backup of the audio in case your level (auto or manual) is incorrect.

Just to be clear: I never use AGC for my audio and that was not the problem I had. I sent the camera back and got a new one by now. The hiss on that one seems slightly less but still not great. I wish I had my AKG mic on hand to test the cam under real world conditions and not only the crappy mic that came with it. When I use my Sennheiser wireless system I don´t have any problems, because I can turn the gain in the camera down to 3. So for me the jury is still out on the pre-amps but they are at least usable and since I like everything else the camera has to offer A LOT, I will keep it.

you can probably use your radio mic on Line level input too, not much chance of noise there and the gain level is set at the transmitter end so other than unexpected overshooting the level there you are safe once you set the input level at the camera end.

I haven't been able to check out the HM100, but I am considering getting one in the very near future.

I use the GY HD100 and often use auto + manual for safety with Rode NTG1 or NTG3 + both these mics put out plenty of level. I have also used the AT 835b with success (less sensitive and less detailed but still a decent budget long shotgun - although at 13" length not for camera mounting).

I have never heard hiss on that cam except back when I was using the "poundland mic" that came with the camera, probably the same one.

First thing I did was replace it with the NTG-1, and last year replaced that with the NTG-3 (which is miles ahead of the NTG1 in capturing detail).

I think don't auto works like agc on these cameras - more like a moderate level of gain with some limiting applied. They will compress loud sounds but can still clip. It was even that way back on the old Sony VX1000. Auto was safe with no pumping of the level in quiet parts.

Generally unless you work in a controlled environment like a studio I would expect ambient sound to be considerably higher than any noise floor on the inputs themselves.

Looks like I'll soon see for myself!
 

... Just won a new/sealed/USA HM100 for $3,000 on eBay! Nobody else bid -- not even a last minute swooper. Makes me a little nervous, actually ...
I should have it Friday or Monday; same with extra batteries, 16 GB cards, cheapo HOYA HMC 46mm haze to protect it just in case I can't stand the boring and not long OR wide enough 39mm-390mm lens and possibly inadequate OIS and yearn for a 16x-20x. I might wait for the replacement for the Canon XH A1S or a stock EX-recording Sony Z series if so.
A smooth Libec H38 head and versatile Bogen 535 legs (10"-67"/75mm/5lbs ... a semi-poor man's Miller DV Solo CF) should come next week, and a nice shotgun of some sort :)

Very good info to know about the NTG-1 vs. NTG-3 shotguns, Thanks! -- I wonder how the Rode NTG-3 compares to a Sennheiser ME66 or maaaybe a cheaper/shorter AT875??

Now, time to install FCE 3.5, then 4.0 upgrade, then Calibrated Q XD Decode, Sony XDCAM EX Clip Browser, and MPEG Streamclip on my poor little unsuspecting white 2006 1.87Ghz Core 2 Duo MacBook with 2GB RAM and a 200GB/7200rpm HD. I should get my external FW400 drive reconnected.
I wonder if my MacBook will be able to cope with anything from my new HM100 ... I'm hoping for at least some short, cuts-only practice and HD viewing on my 20" Cinema.
Tough decision on either a Mac Pro, a MacBook Pro, or probably just a quad-core Hackintosh very soon (Snow Leopard compatibility?!?), I'm sure. And the new Final Cut Studio not long after that?

Shaun -

FYI, I use Final Cut Express 4, and recently upgraded my hardware to one of the current 8-core Mac Pros.

I don't have my HM700 yet (still waiting, better be today or tomorrow, dang-it!), but I've played with some of the available footage online.

You should know that FCE4 does NOT take advantage of all the cores in your machine when rendering. Typically I get CPU usage figures of around 300-400% (of 1600%) when rendering.

Also, FCE4 does NOT offer a 1080p30 sequence preset, so you'll either have to edit in 720p30 if you must have progressive, or edit your 1080 footage in 1080i60 no matter how it was shot.

So I'm planning on upgrading to Final Cut Studio ... hopefully the latest greatest version will be out in the next couple of months and we'll know just how well it uses multiple cores.

Ironically, one function of FCE4 that makes good use of multiple cores is the Log and Transfer panel, where I'll see spikes around 1200% when importing AVCHD. But I'll no longer have to do that as part of my work flow if and when my camera arrives.

Good to know ...
 

So using FCE with the HM100 35Mbps footage is possible, just slow and limited ... Nice!

And there's no image/sound/CODEC quality difference from using FCE over the Studio/Pro version?

Once you get your HM700, what will a brief breakdown of your workflow be?
(I last edited video on a deck to deck Beta SP, S-VHS, 3/4" set ups in 1992-95!!!)

Thanks!

Could I get your email so we can exchange the occasional tip/solution/enhancement and such?

Shaun Walker
shaunotter ( a t ) ya h oo (do t) co m

From ?? to 8-core?
 

What did you upgrade from to your Mac Pro 8-core?

What difference has it made, during what tasks?

Re the Mics mentioned - The NTG-3 is aimed squarely at the venerable 416 and you may find a link to sample files of the 2 side by side in the audio section here on dvinfo.

The NTG-1 is notably crisper an louder than my old at835b. AT seem to do good budget mics (and of course not so budget mics too) and the at827 seems to be a good fit for both quality and size as recommended by those on here who already got their HM100's. Since I have the Rodes I will probably try the NTG1 on the HM when I get one and try out the NTG-3 for size as well although its quite a lot heavier.

>> my mistake the short 7" Audio Technica people mentioned was the AT875(R)

I upgraded from a 1.66 GHz Core Duo Mac Mini with 2GB ram. (NOT a Core 2 Duo). The new machine is a 2.26GHz 8-core Mac Pro (2 Quad-Core Xeons) with 12GB ram (2GB sticks x 6).

Even though FCE4 is effectively only using 2 cores out of 8 on my Mac Pro, in rendering tests I've done, I see a 3X to 4.5X speed improvement over the old Mac Mini. (And something like a 10X improvement for AVCHD log & transfer.)

But if multiprocessing was used to it's fullest potential, I could be seeing a 16X increase... one can hope.

I haven't been able to do tests with Compressor... something got messed up in the transfer of applications from the Mac Mini and Compressor hangs or aborts when run from its own interface. I may need to do a clean install, but that can wait for the next major release of Final Cut.

All things being equal, if you do NOT plan on upgrading from FCE4 for a long period of time, it may be to your advantage to get a 4-core machine at a higher clock speed, rather than an 8-core machine at a lower clock speed.

I'm planning on keeping this machine for a good long while, so I'm betting/hoping that software will be rewritten to take far better advantage of multi-threading, taking more and more advantage of the extra cores in a reasonable time frame.

Any idea about this for without handle/XLRs, as back up ...
 

Mostly as a ultra-compact set up super-short-shotgun, the MKE 400
Sennheiser | MKE 400 Compact Video Camera Shotgun | MKE 400

Can't find info on the AT827 ...

Not that I'm aware of, except for the aforementioned lack of 1080p30 in FCE.

It will probably be a simplification of my current workflow.

Right now I use consumer Canon HF-10/HF-100 cameras for acquisition, which use SDHC cards and record in AVCHD format (1920x1080 in the camera's highest-quality bitrate setting).

The directory tree on the SD card is a real maze, you can't just store the stream files (log & transfer is looking for other metadata), so I archive the cards on my video drive in "Raw Footage Archive" folder, with subfolders by project name, and each "shoot" dated in its own folder, then the exact contents of the SD card. Basically, I'm keeping an exact copy of every SD card, so that it could theoretically be replicated and stuck right back in the camera exactly as it was found. I then do log & transfer imports, as needed, from these archived folders.

But the HM700 needs no Log & Transfer stage, you can drag clips directly to the timeline, so I may wind up storing raw footage under the same folder tree where I store my final cut projects, additional media (logos, stills), etc.

I'm right there with you... I do primarily software development and web design, with a smattering of video production for clients who need it and for local civic affairs. My last "true pro" video full-time work was shot/edited on 3/4", and a few later projects on Beta SP, late 80's into early 90's.

I'm sure I'll spend a good deal of time unlearning the bad habits I've picked up since then. :-)

For example, since most of my current work is for web-publication only, I've started using a little trick where I shoot in 1920x1080, but edit in 1280x720. Any time something is at the periphery of a frame that I want to hide, or any time I want to add a little panning or light zooming effect, I can expand the footage quite a bit (to 150%) without losing any resolution or seeing major artifacts.

But if I intend to do high-quality work for distribution as 1080, none of those cheats will work.

bob at peak dot org
or
sales at sqwarellc dot com

Forecast calls for SNOW ...
 

Snow Leopard OS X 10.6 in the next few months could make life happier on the multi-cores, and add GPU utilization.

Gizmodo - Mac OS 10.6 Snow Leopard Revealed: Multi-Core Optimized, GPU Lovin' OS Upgrade Due In One Year - Snow leopard

Final Cut Pro Studio 3 to allow realtime editing of 1080P H.264 video? | 9 to 5 Mac

...and now back to our regular program...