dBpoweramp - cd ripper software 'compression' setting question
Posted by: Naim727ca on 16 January 2013
Folks, how are things? I have been having fun setting up and playing with my NDX and iPad...awesome toy. I have dBpoweramp version 13.3 to rip my cd collection and when I tried it, there is a 'setting' to compress FLAC flie, it ranges from 0 to 8. The '0' is (fast), '5' is (default) and '8' is best. I am not sure what is ideal to get the best music from this conversion? Please advise from the experts who have tried it. Thanks in advance.
As well, do you know if NDX takes the new USB 3.0 stick?
If you are transcoding to wav in your player it shouldn't matter - I use 5. If you are sending FLAC to your streamer the 0 would probably be best.
If you are transcoding to wav in your player it shouldn't matter - I use 5. If you are sending FLAC to your streamer the 0 would probably be best.
+1
I have tried setting 8, but it takes much longer to encode for minimal reduction in size. 5 seems to be the best compromise.
If you upgrade dbPoweramp to the current v14 you have the option of no compression when ripping to flac. However, this will give you files that are slightly larger than if you ripped to WAV.
I would suggest that you spend a little time and experiment, firstly, can you hear a difference between WAV and flac files when played via the NDX? Some of us can others say not, you need to decide for yourself which applies to you.
If you decide to stick with flac then ripping a few tracks at different compression rates should help you identify the optimum level of compression for you, nobody else can realistically decide this for you.
Also worth consideration is the server used to deliver your files to the NDX, can yours transcribe flac to WAV on-the-fly? This would leave the NDX with less work to do and more time to play your music.
Why not try it? You may find it even more fun.
Regards
Peter
According to the table published at the FLAC homesite here FLAC -5 is the quickest (in overall terms) to decode, then -1, then -3, then -0, then -8. In terms of cpu time alone -0 is quickest, then -1 etc. But these differences are extremely small, 3% or 4%, and in any case vary depending on the type of music as you can see from other data on that site, so encoding at level -0 is no guarantee of the lowest cpu overhead in decoding - sometimes a more compressed file is quicker to decode than a less compressed one.
If you upgrade dbPoweramp to the current v14 you have the option of no compression when ripping to flac. However, this will give you files that are slightly larger than if you ripped to WAV.
I would suggest that you spend a little time and experiment, firstly, can you hear a difference between WAV and flac files when played via the NDX? Some of us can others say not, you need to decide for yourself which applies to you.
If you decide to stick with flac then ripping a few tracks at different compression rates should help you identify the optimum level of compression for you, nobody else can realistically decide this for you.
Also worth consideration is the server used to deliver your files to the NDX, can yours transcribe flac to WAV on-the-fly? This would leave the NDX with less work to do and more time to play your music.
Why not try it? You may find it even more fun.
Regards
Peter
An interesting test but the NDX is decompressing the flac on board. Why not try transcoding both files to wav and then see if you can hear a difference?
Of course you should do whatever works best for you in your system.
Interesting response, in view of your findings I would definitely try ripping to WAV, you may well find that you hear a further improvement. I am not suggesting a massive step up but certainly worthwhile, a free upgrade maybe.
The file size will likely be slightly (very slightly) smaller than the uncompressed flac files for any given track. Alternatively see if your server, if you intend streaming, can transcode flac to WAV. Personally I prefer to rip to WAV, again, if you intend to stream check your server is able to stream WAV.
Good Luck
Peter
If you have dBpoweramp you shouldn't need to re-rip your cds to get a WAV version - dBpoweramp comes with a Music Convertor and a Batch Convertor program which will convert FLACs to WAVs or any lossless format to another with results utterly bit for bit indistinguishable from a rerip. Pretty quick too. If you still want to be really really really sure there is an option to show the checksums to prove the conversions are identical to the rips.
I will try the wav vs. uncompressed FLAC tonight and report back.
Just wondering how does the FLAC converted to WAV on the fly work? Software setup through NAS? I do not have a NAS yet, it is on my thing to buy once I save up some cash. All the budget went to the NDX and now just running through USB stick. These toys are 'expensive' for a work 'joe' like myself.
Quick question on the dBpoweramp, after it ripped each track it provide a number, ie, 7, 8, 9....is that number to tell us the accuracy of the rip? Please advise. First time using this software. Cheers.
The number I think you refer to in dbpoweramp (preceded by ‘Confidence') is the number of rips that the database has recorded that exactly match the checksum of your rip.
Transcoding as we are referring to it here is generally carried out by your UPnP/DLNA Server. This may be on NAS or PC/Mac whatever you use to stream your music. It should be pointed out that not all NAS units with a Server built in are able to transcode. If you decide you need this capability then be careful to ensure any NAS you buy is capable of doing this.
As you are using usb sticks at this time then transcoding is not an option.
If you use a PC and have a network now you can install a UPnP Server such as Asset and try this out, as Asset can transcode on-the-fly. I can’t advise if you use a Mac but others will be able to if you ask.
Peter
Compared between 'FLAC uncompressed' vs. 'WAV' file of the same song, ripped via dBpoweramp v.14.1 - through NDX / 202 / HiCap2, Napsc / 200 / Harbeth C7 = my ears and brian CANNOT tell the difference. I was flipping back and forth between the two formats. Honestly, it is too difficult to judge which is better. Both files ended up with 45 MB each, so in terms of ripped data, it is almost the same. This is my report. Thanks all the folks who provided the advise. I did learn something new. Happy listening.