The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Sound Forge users? A troubling problem...
Old 12th October 2017
Here for the gear

Sound Forge users? A troubling problem...

So I have used Sound Forge for at least a decade now. I have SF Audio Studio 10 and SF Pro 10.

And recently I discovered something very troubling and I'm pretty unhappy about it because I think I'm going to have a lot of further disappointment ahead of me.

Recently I started using a spectrum analyzer to look at the mp3s that I DJ with. I will often use SF and FL Studio to make my own remixes or edits of songs. Sometimes the source is WAV and sometimes it is mp3. Sometimes I will just give older songs a volume boost with the SF limiter.

I began to notice that when I looked at mp3s that were created by Sound Forge, the spectrum analyzer would show that my mp3s were identified as 320kbps yet the frequency graph shows that they are a much lower bit rate! This is beyond bad because it is possible that I have messed up tons of mp3s and songs that I edited and resaved.

So just to make sure you guys understand what I'm saying. When I re save an mp3 in either version of Sound Forge, my spectrum analyzer says that it is in fact 320kbps but the frequency information reads more like it is 192kbps. A lot of information is missing.

So I began to download various freeware editors and even a video editor. Every single one correctly re saves or converts to mp3 and the spectrum analyzer shows a much greater amount of frequency range in the mp3.

Here is the real kicker. In most cases, the file sizes are the same no matter what editor I use.

How on earth is Sound Forge saving my mp3s at the wrong bitrate, yet having the same file size? And if it's the same filesize, how is more frequency information missing that what should be?

I really like Sound Forge and hate to have to start using two editors or going with something else totally.

I'm sick at the thought of how many of my songs might be ruined and a lot of them I won't have unaltered backups for. :(
Old 12th October 2017
So you see a high frequency roll-off?

The frequency range is determined by the sampling frequency, not the compression rate. It's possible though that they roll off high frequencies by default to suppress audible artifacts?
Old 12th October 2017
Here for the gear

It is doing it to WAV files to. A 320kbps mp3 used to reach the 20khz range. Now they are saving as only reaching the 15khz range.
Old 12th October 2017
A few things:

First, the term bitrate is a measure of size per time, nothing more. Of course, properly made mp3s tend to sound better the more sound data we leave in them -- but not all codecs [encoder/decoder software sets] are created equally, even in the 'mp3' arena.

My version of Soundforge is old and only came with a 'trial' version of the Fraunhofer codec, good for a limited number of encodings. So I don't know what's in the current version, but I would suspect it's Fraunhofer -- particularly now that the Sonic Foundry software that used to be owned by Sony is now owned by German soft-boxer, Magix.

(I'm among the substantial cohort who tend to like the LAME encoder better than Fraunhofer. But because of licensing issues, most mainstream commercial software that includes mp3 encoding uses Fraunhofer.)

It's also important to remember that all encoders (we're more or less down to the 'official' Fraunhofer and the open source LAME) have different modes of operation, most importantly for this conversation, a tradeoff between speed of processing versus quality of processing.

Me, I always use LAME's slow/high-quality setting because it's just not that much slower and sound quality matters more to me than one-time processing waits. (You typically have to use a 'wrapper' or command line switches in 'terminal' mode [DOS, so to speak, on Win machines].) The 'perceptual encoding' algorithms on the second pass will go back over the already encoded file

One last -- but maybe crucial point -- never, ever, ever use an mp3, AAC, Ogg, or other lossy-compressed sound file for MORE processing and then re-encode it! (If you can possibly avoid it.) ONLY use such files for playback. Always use uncompressed wav or AIFF files for processing, editing, etc.)

Passing an already lossy file through more lossy encoding is often called 'transcoding' and it is a sure prescription for gnarled, mangled high end and more -- and it will, indeed, tend to make you think of really crappy, low-bitrate lossy encoding.

If you need to archive or save in a smaller format in between working on a given sound file, use the truly lossless FLAC (or Apple Lossless in the Mac/iOS world if you prefer).
Old 9th June 2018
Here for the gear

Sorry to dig up this thread again, but did this ever get resolved?

I have been a SF user since 4.5c back in 1997 and do all my MP3's and WAV mastering on it..

However, as over the years I use a traditional DAW like Ableton Live 9.7, and Logic Pro X, I always found that it was so easy and destructive (what im used to) to edit, top tail master tracks in SF (and im on a Mac, and hated the SF mac version, so have to keep a PC around for the older SF versions)

BUT just lately using Logic Pro, and Ableton, I did find on Headphones that the mixes from SF didn't sound the same (WAV 44.1khz) as they did coming from Logic or Ableton?

I did think about it, but realised that there was thousands of DAW comparisons, saying that one is better than the other, and a huge debate starts that they all null against each other.

But is there something going on here that SF does to mp3 and WAV files once re-saved in SF?

SF is great dont get me wrong, but I still use an old version, and really do want to move over to something like Reaper (which is similar in many ways) but its not a proper editor..
Old 9th June 2018
I would strongly doubt that WAV files rendered from SF and WAV files rendered in the same format with no additional processing from other editors or DAWs would differ significantly.

However, one needs to be sure that there is no other processing. Once any sort of processing is invoked, the potential for differences in that processing, in dithering, etc, can enter into it -- and it's not guaranteed that those processes will be identical.

Also, with regard to Mp3s, obviously there can be profound differences when using different bitrate settings -- but something a lot of folks appear to not understand, mp3 encoders like the 'official' Fraunhofer encoder or the open-source LAME encoder (which many folks prefer) have a multitude of command line settings that are typically handled by a 'wrapper' of some kind, sometimes a standalone app (like WinLAME), or by the host editor or DAW. (Good editors and DAWs will let you define your own call parameters for customizing mp3 settings.)

In fact, there can be noticeable differences between the 'fast' processing that many default to and 'slow,' higher quality processing -- particularly, for instance, when moving between the Fraunhofer encoder at its fast/low quality default and the LAME encoder set up for the highest quality/slowest processing.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.

Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump