Freelancer soundtrack remaster.
-
I’ve a solution! I will do an old idea. Creating an artificial ear to have the most flat rig by measuring my headphones with some tweaks and after do the remaster with a combination of spectrum and my ears. While the result isn’t perfect I have to make it better.
I had 3 opinions but it’s not enough, I want to create something better to show it in other forum and make a mod db page, some remasters are lacking volume so it isn’t right even if using a compressor which totally disable the original volume. And even you have a good knowledge in mastering, there is always things that have to be perfected every time you use another file. Sadly, in other terms I have to reboot my remastering cause of the fact that my sound equipment is not fully ready to make it, even if it’s good it could be better especially in bass range that I can’t measure properly…
I don’t say that’s bad, I say it’s perfectible. The main problem is in the trebles because I don’t edit the bass.
In this case I will put some efforts on keeping Freelancer spirite and improve it.
Edit : Crap! I dunno what to do now… Should I prefer the original mastering or continu what I’m doing? I’m lost in the see…
-
I have taken some time to listen and compare both and believe in the original version the filtering is intended to create a certain atmosphere. If you listen to the different regions, there is quite a difference how it sounds. So I personally find your remaster worse, because e.g. for Bretonia it doesn’t sound as grim anymore.
If I am right that doesn’t mean the whole idea is bad, but maybe we need a different approach. I also really appreciate your enthusiasm, so don’t get me wrong
Also maybe you can explain what exactly you are doing. For a remaster I would have assumed that you need all the channels for each instrument to adapt the mixing. Since you don’t have that it is probably much harder to get good results, right?
-
Before going to sleep, I have got a crazy idea, looking at vanilla spectrum. And I have equalized the CD soundtrack from the informations I have got from narrowed sampling rate vanilla ost and you will not believe me… That’s it! The result reflect totally what it should be! Just an example, m13 final at 10 kHz vanilla -78 dB, CD OST -81 dB, elevation of 3 dB of the CD OST and in the trebles, the instruments are really accurate and the main ambiance remains… WTH! for 3 dB! I know that 3 dB double the sound volume but for 3 dB everything is better. Vanilla OST despite its low sampling rate is far better mastered than CD OST. Another example, nomad lair battle, with the informations got in the vanilla one, the ambiance has been restored in the CD OST and once again, that was exactly what the soundtrack was intended to be! Unlike Starcraft CD OST, the master of FL CD OST was probably done with a high trebled speakers or headphones or they tried to make things like me but with their speakers… 3 dB… -_- I’ve a reason to be enraged…
So right now, vanilla OST will be the reference and not Pink Noise, I think it’s the best approche. I have taken the buoy, not a big one but a high quality one, I’m not any longer lost in the see.
-
@ Schmack : You’re right! That was exactly what I thought after some thoughts. Things need to be as close as possible to the original ambiance and artists choices. If it remove the goal of the track it ruine it. The only thing to change is to make the remaster as close as possible to the vanilla tracks. Now I know where I go.
-
Freelancer remaster comparison.
Content :
- CD OST
- Pink noised master = 1st remaster
- FL remaster = CD remaster based on vanilla master.
I create a poll to see which is the best (and don’t forget to use headphones!)
-
Why no one else wants to reply to this topic? And votes for one of the propositions? Are you waiting that I do the work? I would like to have some opinions on the comparison, I want to give the best material for the community so once again, I would like to have your opinion. :?
-
In my case I currently don’t have much time plus want to give you a proper opinion. So it will take a few days to listen to it. Also many people don’t have that well trained ears. I can tell you whether a music file was lossy compressed and usually also which compression was chosen (if it is one of the common used ones), but most people will not hear a difference.
Currently I am not decided, because you seem to be in between the mp3 compressed files and the cd sound track. Will take some time to decide.
-
OK! When you have the time, make your decision.
You can compare them with speakers too because I just changed the trebles.
-
Freestalker.fr wrote:
Freelancer remaster comparison.Content :
- CD OST
- Pink noised master = 1st remaster
- FL remaster = CD remaster based on vanilla master.
I create a poll to see which is the best (and don’t forget to use headphones!)
The FL remaster is brighter than the OST, makes it sounds more catchy and more accurate.
The pink noised remaster lacks of lows and it is more difficult to feel the ambience and the deepness.
I think you should keep your remaster, for me it sounds good !
And sorry for not replying, I’m not coming so often here.
EDIT : I used an Audio Technica ATH-M50x headphone.
-
Thank you for sharing your opinion. I have taken it into account.
Wow! I will have a bunch of work to do! I shouldn’t have watched the file number… I have to find some energy to make it. (I think after some done files, it should recharge batteries. ^^ ).
Now every day I will make a report. It could allow me to make a certain number of file per day.
PS : ATH-M50x, rather a bassy one but not as bassy as certain measurement shows, Audio Technica is renowned for their headphone quality in audiophile world. It’s a nice choice though, despite my linearity fanaticism, I already owned bassy headphone (Superlux HD 681) and the sound remained clear, and also the treble was piercing but every time I looked at a meas site to equalize them… ‘~’ But there is a website called Golden Ears which shows quite accurate curves. Make sure that curves are green and blue (I didn’t try to test them but I trust their Head and Torso Simulator. I have to buy a HD 668B to test.)
-
I wasn’t quite satisfied with the sound, so here is my take on it: Link
I used my hearing to correct the bass, mid-tones and trebles for each file individually. My goal was to get a rather clean sound. I realize this is a matter of opinion
Also I only have a Sennheiser PC161 headset, but I used my X-FI Titanium HD in audio creation mode with bit-matched playback and ASIO driver for listening to get as close to the original sound as I could.
This is a rather slow process but if you guys like it I can do more. Maybe it at least inspires you
-
I made some comparison and your’s is more accurate in the high mids from 1k to 6k, as for mine, that was before I made a refined process, that wasv one of my first files, so it isn’t the case for all the files of the comparison pack. I put a screenshot of the spectrum of vanilla, schmack one and my one.
We are both right. But I want to make it! Schmack, you’ve already done a lot for this community, so could you let me do that? please? It makes me happy and allow me to work in one of my beloved domain. If I struggle with this task I will surely ask for some help but for now it seems that I prefer to do this myself. Do not believe it’s against you, I really like you, it’s just to force me to do a project and you’ve a lot of work to do with your OpenGL wrapper, it’s a long and tough task to do, concentrate you on this, you’ve no time to lose in this, you’re not Atlas!
-
Schmackbolzen wrote:
I wasn’t quite satisfied with the sound, so here is my take on it: LinkI used my hearing to correct the bass, mid-tones and trebles for each file individually. My goal was to get a rather clean sound. I realize this is a matter of opinion
Also I only have a Sennheiser PC161 headset, but I used my X-FI Titanium HD in audio creation mode with bit-matched playback and ASIO driver for listening to get as close to the original sound as I could.
This is a rather slow process but if you guys like it I can do more. Maybe it at least inspires you
Nice and crisp dude.
Maybe do some remixes with new samples? -
@Freestalker: Don’t worry, I won’t take it personal and as I already stated I appreciate your enthusiasm Also having my own projects I understand how you feel about this.
I just wanted to give some constructive criticism and thought the best way to do that is to let you listen to how I think it should sound. Also I have never done this before, so it was quite fun and a nice diversion But don’t be surprised: Audio processing is very similar to image processing, so I have all the theoretical background which is needed, plus my minor is analog and digital electronics, that also helps a lot@Bolte: Glad you like them! Here is another one, but I don’t think I will do more as it is rather tedious (also this one has really strange original mixing, which made it more difficult): Link . Maybe if I get specific requests and if I also like the track I will give it another try.
In general I listen to a track the next day before I decide whether it is good enough, since I found, like with most of your own stuff, it helps to get some distance to it and evaluate it with a fresh mindset. This is how I usually get the best results.
-
Freestalker, you are doing good, don’t stop because someone else stepped in and did something too. This is how progression is done, one person doing something, then another pushing the limits, inspiring someone else to make more improvements.
We would have no technology such as we have today without this train of progression, keep at it
-
Thank you! Bolt and Schmack.
Schmack, your master in the lows is nice and reflect totally what it should be. But in the highs the sound lacks of punch and remind me too much the low quality master of the CD soundtrack, you have to remind that the CD OST is here to give the data, if the treble line is not like the vanilla OST, it cannot make the spatialisation correct and make the sound compressed.
Kusari is normally cold and spacious, it allows you to breath like when you are under the see, the sound design tends to remind you of documentaries music. It’s something between calmness (zen mood) and brightness (sparkling blue everywhere so higher trebles).
I concede that the background noise is changing and the sound become noisy. There is a VST called Voxengo Redunoise that can learn the noise via the spectrum for that, it’s incredible!
-
I know that the trebles are not that powerful, because I did not like the sound of it when I increased them too much. So there is only minimal gain in all the tracks I processed. Maybe it is a matter of taste or you find a way to make it not sound so bad
I am curious with what solution you’ll come up with and how it will sound in the end, that is for sure! -
I’ve found the way, I just equalize before 10kHz and after it’s flat. For example I boost 10 kHz for about 6 dB then higher frequency have the same boost. I’m doing a sort of shelving filter with a corner frequency of 10k and not a peaking filter with a center frequency at 20k. We have no clue of what it should be after 10k, so I let them with the same slope but I have boosted them and the sound is not too harsh to hear, the result is nice, despite it’s not as fantastic as my first days on this remaster, no more thrills now.
PS : Omicron musics get severely hurt by the mastering, and now, it’s really clear after my remaster.
-
When you guys come up with a finished touched up remaster, that is complete. I will put them into a patch. That is if you want.
Since I’m putting together one complete patch to fix the common problems folk have running the game to bring the source up to date, instead of bits scattered around the web.
It’ll have a complete launcher with auto-update capability for the main source data files and for mods.
It’ll be put into a git source on a repo for everyone to have a go at forking. Every pull request to update the source will go through voting process until there is a majority, only then will it be accepted into the source.
So basically everyone will get their chance to put their mark into Freelancer.
-
But personally, I like Schmackbolzens versions. The ones from #24 really highlight some underscores I hadn’t really noticed before because of the quality I’ve heard them at.
For instance in the Kusari one there’s some sounds I hadn’t realised were there because they were very clean and the sounds were highlighted more if that makes sense.
That’s what I want to see (or rather hear) from this.
Not that I’m saying what you are doing is bad my friend.