I posted this elsewhere but thought the Zoom thread was the right place. Based on all that's written here, I just bought a Zoom H4.

My previous recording rig was a cheap mic plugged into a creative soundblaster external plugged into an aging (8 year old) laptop wirelessly connected into my home LAN. I'd record using audacity then use my desktop computer to do any work with it (normalize, convert to MP3, etc). This arrangement was really painful for a lot of reasons: having to start the laptop up (takes forever since it's so old) whenever I want to record; the piano is downstairs and the desktop is upstairs; wires all over the place.

More importantly, the final result doesn't sound very good. It's one thing to have a cheap piano and have a recording sound like cr*p but when you've got a good one it's really disappointing when you finally get a good "take" and the results sound nothing like the piano.

I'm looking forward to getting the Zoom! I don't expect "professional" level results from it but, from what I've read here, what it lacks in quality it more than makes up in convenience. Besides, my playing doesn't merit a "professional" recording rig and I expect that this is yet another case where the equipment is better than the operator!

I stumbled on this review of the Zoom that explains the various functions pretty well:

Review of the Zoom

Since this thread came up to the top again, here's something to experiment with (I haven't tried it yet). LL's frustration with WAV to MP3 conversion reminded me. This suggestion has to do with "recording format" :

On the outside of the H4 are "track" buttons that let you quickly choose a format. The only MP3 option is at a sampling rate of 44.1KHz.

But, the H4 has many other options (higher sampling rates) for direct recording to MP3. See
pp's 34-35 of the H4 manual for instructions on setting the "recording format."

For most of us, it's pretty much been a given to record WAV (very high quality), transfer to computer, then convert WAV to MP3. We think this gives a better MP3 product than the default H4's MP2 recording format (?).

But, perhaps, if MP3 is *really* what you want, it would be better to let the H4 record directly to MP3 -- but choose a sampling rate of, say, 256Kbps. This might be a better MP3 product than what is produced by RiverPast, Audacity, or LAME ???

I'm too swamped with work and spring chores to experiment right now, but perhaps somebody else could try and compare.

Cheers. --G

Ragtime, careful not to confuse sampling rates with bit rates. For MP3, I believe that the sampling rate on the Zoom is fixed at 44.1khz, however the bitrate can vary as low as 48kbps (yuk!) all the way to 320kpbs or VBR (variable bit rate). The bitrate controls how aggressively the audio will be compressed.

As to your main point, there certainly can be a difference in the quality of codecs doing the wav->mp3 conversion. Which is better is hard to know without testing, and even then it may turn out that different codecs have different strengths/weaknesses. Also, if many people are using iTunes, going from wav->aac might actually be a better solution.

Keep in mind that one of the greatest advantages of recording in wav is that you're free to compress to a particular bitrate and then choose another if you need to (or an entirely different scheme like aac). If your original recording is already compressed, you can never get "better". Lastly, don't forget the golden rule that if you'll be doing almost ANY post processing of the audio, use wav even if the final product will be mp3 to avoid the cumulative effects of lossy compression.

Thanks for the note.

"Bit size" is the accuracy of every sample (16,32, 64, 128, ... bits). Consider this: measure the width of your Ab key. In millimeters, you might get 12.7 or 12.069, or 12.0692 -- each being a more precise way of expressing exactly the same thing.

"Rate" is how often a measurement is taken (44.1 measurements/second, 96 measurements/second). (It's unlikely that you would measure the width of your Ab key more than once a week).

Digital music quality is affected by both the size of the sample, and the frequency at which the sample is taken. I think the H4 has a fixed bit "size" and a variable sample "rate".

In my opinion (experts otherwise please weigh in) *rate* has much more influence on our recorded music quality than the *size* of each particular measurement sample.

Let's test. ??

Sorry Ragtime, but you're still confusing bit rate and sample rate.

So we'll break it down to a two step process, the front end and the back end. In the front end (which is independent of output format (wav or mp3 or aac), sound is "sampled" at a particluar sampling rate. So if the sampling rate is 44.1khz, that means that 44100/sec the amplitude of the sound signal is measured. Now that value is stored in some number of bits, the bit size (aka sample size). The two most common bit sizes are 16bit and 24bit. 16bit is most commonly known as the size that the cd redbook standard uses. 24bit is what dvd audio uses. 24bit audio simply means that the amplitude can be approximated with more resolution that 16bit (~16mil unique values vs ~65000).

OK, so sound is being sampled at some rate and stored in bit sized values. If we do nothing else then we basically have a raw wav file. Now that wave file is large,

44100 samples/sec * 16bits/sample * 2 channels == ~1400k bits or ~176k Bytes per second of music, as a bit rate, this is expressed as 1400kbps.

So we apply an audio compression algorithm to the music to shrink those file sizes. The way most if not all the audio codecs work is to choose a bit rate that is lower than the 1400kbps of the uncompressed file. A common bitrate for mp3's is 128kbps, a pretty significant savings over the raw bit rate. So a 1 minute raw wav will be about 10MB while the mp3 will be about 1/10th the size or 1MB. Notice that we can choose whatever bit rate to compress at, but that has absolutely no effect on the fact that our front end sampled at 44.1khz at 16bits/sample.

More to follow ...

So compression doesn´t involve a better encoding of redundancy?

There are lossless compression algorithms that just use redundancy to reduce filesize/bitrate, but they don't compress by anything like as much as the lossy ones such as mp3 - perhaps a factor of 2 instead of a factor of 10.

Best wishes,
Matthew

Ah..

Thanks. I will keep that in mind.

For an H4 carrying case, I found the Zebracase Z12. Online music stores have them for about $30. The foam is perforated in a grid so you can customize the cutouts by removing squares. I can store the recorder, AC adapter, memory cards, manual, wind screen, USB cable, and tripod mount (not shown).

Got it. Thanks! My suggestion is to try the H4's native MP3 options (using the menu, not the front panel buttons) and see if the result is better than a computer program's WAV to MP3.

I agree that will be interesting, though my gut feel is that the convenience of not having to go through the extra compression step will outweigh any audible differences.

I think the setup should use music playing from a decent stereo system or someone who has a player piano or good digital. I think it would be far to tricky to get someone who could play a piece almost exactly twice in a row (1 mp3 and 1 wav) let alone three or more (x mp3s at different bit rates and 1 wav). (Ideal scenario would be to have 4 Zooms recording the same live performance ) Choose say 3 different bit rates (128 because it's common and about as small as you are going to get without seriously compromising the music, 256 a nice jump in quality, and VBR which is about as good as an mp3 is going to get) and 1 wav recording.

To make life easier, the wav could be posted somewhere and other volunteers could choose a codec to use and encode at the same bit rates. Then the files could be posted and people allowed to do comparison tests. For good measure I'd recommend throwing in some AAC encoded files due to the ubiqitous nature of itunes and maybe FLAC and AAC Lossless as examples of lossless compression.

First step is the base recordings. I'll be happy to volunteer, however my schedule is packed for the next several weeks (end of school, recitals, work, ...) so it may take me a while to actually get to it.

Then again, like I said, maybe this is overkill when the answer is likely, use internal mp3 encoding on the Zoom if you're just going to distribute it on the web, it's close enough. But then again, like they say, "if it's worth doing, it's worth over doing"

I just encountered something with my Zoom that I thought might be worth resurrecting the mega-thread before.

Today I went to record a piece, but when I listened to it the sound levels were very very low... not at all what I've been used to getting when I use auto gain. So I went to the "input menu" and discovered that my auto-gain had been turned off. I then remembered that I had been using batteries but had to replace them. So I am wondering if switching the batteries out somehow wiped out my settings.

Anyway, that's something y'all might want to watch out for, and if you're suddenly not getting sound levels you're used to, check to make sure your auto-gain is still on.

I have a question I have not been able to find the answer to. Does anyone know how to set the date and time on the Zoom?

I just saw an ad for the H2 in the June issue of Recording magazine. It indicated that the H2 has 3 directional mics built in which function as a mid-side stereo array. Which is quite unusual. It sounds similar in concept to the Paia kit mic which uses 3 omni's instead. There was no indication in the ad whether it took external mic connections or not. Looked cute, though. Kind of like a cell phone.

Howard

EDIT: here's a link.. http://www.samsontech.com/products/productpage.cfm?prodID=1916
... looks like there is an external mic-in

Hi all,

I received a Zoom H4 yesterday, and you know, I have the mrs-8 as well. It's not as portable, but it sounds better, to my ears. I'm returning the H4. I thought it would be handy, but I am demanding better sound from my unit. I didn't find it worth the money. I messed with ALL the controls and really got in there, but just wasn't that pleased. I really wanted to hear some of these samples but there weren't many to listen to as they have all been removed. Anyway, just my .02 cents. I don't think the mrs-8 is all that great either, but I have my real studio with an echo layla 24/96 at my boyfriends house. My piano is at home, and a bear to mic, it's an upright, and I've always wanted to record it. So I had the thing tuned, got the unit and started recording. Another plus to that unit (at the same price) is that it will do 8 tracks, with each track having 10 virtual takes, so you can really have 80 tracks to choose from. All the effects are there, and it's just a fun little unit. Harder to figure out than the H4 though. It would have been handy to use this at a jamfest I'll be attending in May, but I can't keep it just for that. My mrs-8 has a good built in condensor mic too, and runs on either batteries or power supply. I recommend using a pre instead of the input pres on the unit though. I would recommend that with the H4 too. I find both units have a lot of noise. I wish I had a recording to demonstrate, but unfortunately I don't at this time, but I'll try to get something here sometime. I like mine as it's SO MUCH EASIER to lay down a vocal to my piano tracks. It only has two inputs or I'd just do them all together. Anyway, rant over, and I'm really glad I found this forum.

P.S. My piano is an old Steinway upright. Could use some cosmetic work, but sounds okay. Pretty dark sounding, though. Hard to record. Must use eq. Okay, have a great day!

I decided to keep the unit, because it records my boyfriend's guitar really well. I'll use the mrs8 for vocals.