Floatplane DOES NOT provide users with better audio quality

[AHPanda]

5 minutes ago, starsmine said:

your understanding of the Phoebus cartel is wrong. It was not purposely downgraded, it was more efficient

What is better
a 10 cent lamp that makes 1k lumens at 60 w for 1000 hours or
a 10 cent lamp that makes 1k lumens at 80W for 2500 hours or
two 10 cent lamps that make 500 lumens each at 40w that last 2500 hours?

for your wallet, it is STRONGLY the first one. it uses 150 kw/hours (60*2.5 to make it last 2500) while the inefficient long-lasting lamps use 50kw/hours more. 

The cartel lowered operational costs and worked to standardize the life expectancy of light bulbs at 1,000 hours^[6] (down from 2,500 hours),^[6] while raising prices without fear of competition.

Try to twist or turn it, they did it for their bottom line, not the consumers'

[rcmaehl]

12 minutes ago, AHPanda said:

And there's the problem. That line says everything you need to know about consumers these days.

Just to go back in time a bit. Back in te 30's, several lightbulb factories set up an arrangement to purposely downgrade the quality of their product so they wouldn't last longer than x amount of hours. The public backlash was "it is what it is". That was +90 years ago, nothing changed.

 

[adm0n]

1 minute ago, AHPanda said:

The cartel lowered operational costs and worked to standardize the life expectancy of light bulbs at 1,000 hours^[6] (down from 2,500 hours),^[6] while raising prices without fear of competition.

Try to twist or turn it, they did it for their bottom line, not the consumers'

In most cases it is actually less expensive to buy more light bulbs, if they cost less to run.

[Divritenis]

 

Am I missing something here? Why is this person comparing 1080p vs 4k? 

[yoc]

13 minutes ago, Benergy said:

I made no such claim.

 

 

Why not send this to Floatplane support and ask why your audio quality isn't what you expected?

It could be as simple as an editor making a mistake when exporting the video, and I am sure they would be happy to sort out any issues you are having.

 

 

Uh... Why not share the findings publicly? Really, what's wrong with it?
I think that is good to do so, at least is not one-sided discussion.

[yoc]

2 minutes ago, rcmaehl said:

 

Great channel! I fully recommend it.

[TimedPing]

1 minute ago, Jazmak said:

because Nvidia was successfully sued and had to pay $30 to every affected US customer over it.

That's exactly why I using GTX 970 fiasco as an example, if people is feeling getting ripped off or etc, just pursue legal action or asked explanation from the floatplane itself before labeling it "Another lie exposed" .

 

3 minutes ago, Jazmak said:

I don't know where you're getting your "considered a well investment" metrics from

I dunno? from both the software support from NVIDIA and compatibility from many 3d party software vendor like adobe? We talking about 2014 card here, that only massively decommissioned by many user just around in the end of 2010's because how good it is, especially when the GPU cryptomining hit. In 2022, I help my client decommissioning many of their workstation rocking GTX 970 (there's around 34 of them) from early 2015. They just recently upgraded mainly because the power consumption cost instead of the hardware performance itself, especially they heavily using CUDA and NVENC in their workflow.

[Vexenar]

4 minutes ago, Divritenis said:

 

Am I missing something here? Why is this person comparing 1080p vs 4k? 

I think he had a typo as his sentence after that focusses on Youtube vs. Floatplane 4K quality. 

[Gus Pyles]

This VAT stuff is why you don't roll your own payment solution. It's a pretty big oopsie doopsie by FP, and they should maybe get sent a reminder of the appropriate EU regulation.

[Benergy]

3 minutes ago, yoc said:

Uh... Why not share the findings publicly? Really, what's wrong with it?
I think that is good to do so, at least is not one-sided discussion.

If support immediately responded with "woops my bad, we will fix this immediately", then there is no need for this post. 

 

There doesn't need to be a 'discussion' unless they don't fix the issue.

[starsmine]

lmao
 

Quote

At least to my eyes, Floatplane 1080p does look better than YouTube 1080p. But YouTube 4k looks leagues better than Floatplane 1080p. For the 6% of you that pay the $10/month, your 4k video quality is slightly better than YouTubes 4k I would say. However not by much.

Dude says Floatplane 1080p > youtube 1080p
youtubte 4k > Floatplane 1080p
Floatplane 4k > youtube 4k
Clearly, A lie when this line states the videos are better on Floatplane

ok.

Why did they say LIE? When they could have just said discrepancy. OP is being inflammatory for the sake of being inflammatory, not because he has a smoking gun catching anyone in a lie.

In terms of AUDIO quality, reddit OP themselves found 320kbp/s audio in some of the floatplane videos. AKA what they are finding are what was uploaded. 

Youtube reencodes shit for standardization. Floatplane may just be taking the uploaded audio and going "ok seems good to me". And that is why there are some 96k audio files there you don't find on youtube. 

 

For SPOKEN videos, not music audio, 96k is MORE then enough. You do not go higher if you do not need to (like if your video had music)

[yoc]

4 minutes ago, Benergy said:

If support immediately responded with "woops my bad, we will fix this immediately", then there is no need for this post. 

 

There doesn't need to be a 'discussion' unless they don't fix the issue.

If the findings are true (nobody can be 100% sure yet), maybe a "we will fix this immediately" is not enough since they have been publicly stating that the quality is better. Accountability is very important, one cannot make false claims to millions like that.

[saintlouisbagels]

21 minutes ago, Divritenis said:

 

Am I missing something here? Why is this person comparing 1080p vs 4k? 

 

17 minutes ago, Vexenar said:

I think he had a typo as his sentence after that focusses on Youtube vs. Floatplane 4K quality. 

In very old WAN Shows, Linus would talk up the quality of Floatplane's video quality being so good that FP's 1080p looked better than YouTube's 4K. It's possible it may not be true anymore depending on whatever changes been made on either side.

 

Linus wasn't exaggerating though. Bitrate makes a huge difference whereas resolution is just the number of pixels. I can record identical video content at high bitrate 1080p and low bitrate 4K on my mirrorless camera and obviously the 1080p video would look better. I could also just work with a 4K master file and play around with settings so that the 1080p export is better than the 4K export.

 

- - - - - - -

 

Also the title of this Topic and the title of the subreddit post are stupid as hell.  Sensationalist titles to farm karma and get people riled up.

[Benergy]

1 minute ago, saintlouisbagels said:

 

In very old WAN Shows, Linus would talk up the quality of Floatplane's video quality being so good that FP's 1080p looked better than YouTube's 4K. It's possible it may not be true anymore depending on whatever changes been made on either side.

 

Linus wasn't exaggerating though. Bitrate makes a huge difference whereas resolution is just the number of pixels. I can record identical video content at high bitrate 1080p and low bitrate 4K on my mirrorless camera and obviously the 1080p video would look better. I could also just work with a 4K master file and play around with settings so that the 1080p export is better than the 4K export.

Yes - it was only recently that YouTube introduced the 1080p 'High Bitrate' option.

 

What Linus said about this in the past may have been true at the time but may no longer be true.

[geomac]

This is a whole can of worms - as someone who's going through the video encoding and compression rabbit hole right now with Bluray archiving and Plex streaming.

 

The short answer is it's very complicated with lots of variables, exponentially so beyond my knowledge due to the CDN component, just look at the comments on the reddit post and people are already criticizing their approach of comparing the two different codecs by bitrate and CRF. The easiest approach would probably be a side-by-side comparison of the youtube player and floatplane player, at the location you are because it's entirely possible that the bitrate is optimized based on your connection and bandwidth to the CDN. (Bandwidth to the CDN isn't competely correlated to your ISP bandwidth, you can be in West Australia with a Fiber connection and unable to reach the CDN.)

 

Even the criticism of not using more efficient codecs such as AV1 or HEVC is also misinformed. These codecs require either newer HW encoders/decoders that support it or compute expensive software encoders/decoders.

 

That being said - if you're not getting a value from a service, then don't pay for it.

 

[Sorenson]

1 hour ago, AHPanda said:

I'm no audio/videophile so I can't say you're right or wrong, but it is an interesting point to bring up. But there are still other issues in the topic like VAT not being properly charged in EU (which is mandatory) so it's possible they're dodging taxes.

They don't have Nexus in the EU since they operate out of Canada without a branch/warehouses in the EU. I'm not 100% sure about European tax laws as I'm in the US, but at least for us, this would allow them to avoid many US taxes. 

[RamblingPenguin]

I just had a brief look into this, I compared the FP 4k download of their apology video (the only one where I can get the exact same time stamp on both FP and YT), and looking at the frames the light gradient is way more blocky in the YT 4k version compared to the FP 4k version. I can't comment on the audio part yet.

[AHPanda]

3 minutes ago, geomac said:

 

...That being said - if you're not getting a value from a service, then don't pay for it.

 

I agree, but the issue is that people didn't know this so they paid for it anyway.

[saintlouisbagels]

2 minutes ago, RamblingPenguin said:

I just had a brief look into this, I compared the FP 4k download of their apology video (the only one where I can get the exact same time stamp on both FP and YT), and looking at the frames the light gradient is way more blocky in the YT 4k version compared to the FP 4k version. I can't comment on the audio part yet.

Use your ears...?

[CodeSlinger]

So interestingly I tried this on some videos and they definitely seem to have reduced the quality since mid 2020. and further reduced recently

2 channel super fun videos from 2020:

Video bitrate: 5600 kb/s & 6027 kb/s
Audio bitrate: 140 kb/s 

2 most recent channel super fun videos:

Video bitrate: 2400 kb/s & 4344 kb/s
Audio bitrate: 99 kb/s & 102 kb/s

2 recent short circuit videos:

Video bitrate: 1504 kb/s & 1428 kb/s
Audio bitrate: 82 kb/s & 84 kb/s

Which means that this isn't a service side change but rather an upload quality change, I would bet that something changes in their editing pipeline that lowered the quality of the videos being uploaded in 1080p. To paint this as a "LINUS AND LUKE LIED AGAIN EEEEE" seems a bit reactive, it could be a mistake or an adjustment they should have communicated better. But certainly floatplane is still servicing high bitrate 1080p if the upload was provided at that level.

[geomac]

1 minute ago, CodeSlinger said:

So interestingly I tried this on some videos and they definitely seem to have reduced the quality since mid 2020. and further reduced recently

2 channel super fun videos from 2020:

Video bitrate: 5600 kb/s & 6027 kb/s
Audio bitrate: 140 kb/s 

2 most recent channel super fun videos:

Video bitrate: 2400 kb/s & 4344 kb/s
Audio bitrate: 99 kb/s & 102 kb/s

2 recent short circuit videos:

Video bitrate: 1504 kb/s & 1428 kb/s
Audio bitrate: 82 kb/s & 84 kb/s

Which means that this isn't a service side change but rather an upload quality change, I would bet that something changes in their editing pipeline that lowered the quality of the videos being uploaded in 1080p. To paint this as a "LINUS AND LUKE LIED AGAIN EEEEE" seems a bit reactive, it could be a mistake or an adjustment they should have communicated better. But certainly floatplane is still servicing high bitrate 1080p if the upload was provided at that level.

Is it possible for you to download the old CSF videos again and compare?

 

Verify that you're still getting the same bitrates.

[AHPanda]

3 minutes ago, CodeSlinger said:

So interestingly I tried this on some videos and they definitely seem to have reduced the quality since mid 2020. and further reduced recently

2 channel super fun videos from 2020:

Video bitrate: 5600 kb/s & 6027 kb/s
Audio bitrate: 140 kb/s 

2 most recent channel super fun videos:

Video bitrate: 2400 kb/s & 4344 kb/s
Audio bitrate: 99 kb/s & 102 kb/s

2 recent short circuit videos:

Video bitrate: 1504 kb/s & 1428 kb/s
Audio bitrate: 82 kb/s & 84 kb/s

Which means that this isn't a service side change but rather an upload quality change, I would bet that something changes in their editing pipeline that lowered the quality of the videos being uploaded in 1080p. To paint this as a "LINUS AND LUKE LIED AGAIN EEEEE" seems a bit reactive, it could be a mistake or an adjustment they should have communicated better. But certainly floatplane is still servicing high bitrate 1080p if the upload was provided at that level.

It is reactive, because it's true. Even if this is a genuine change on their part it doesn't change the fact consumers pay for better quality. Thus, it needs to be adressed.

[CodeSlinger]

2 minutes ago, geomac said:

Is it possible for you to download the old CSF videos again and compare?

 

Verify that you're still getting the same bitrates.

I just downloaded all 6 videos 5 minutes ago. This isn't on old downloads.

[geomac]

1 minute ago, CodeSlinger said:

I just downloaded all 6 videos 5 minutes ago. This isn't on old downloads.

Thank you for the clarification. The issue then is with they're editting pipeline as you described. The CDN and player seem to have the same capabilities.

[DoctaJones81]

I've not looked at any of the info directly myself, but anything with a variable bit rate is likely to have a target, but then the final reported bitrate is going to be very tied to the actual content of the video. In other words, the content of a Short Circuit video vs a Super Fun video are vastly different. The first is primarily a single individual usually seated and not much in the frame of video is moving. Whereas there's a lot more full frame of movement in a Super Fun video. Long story short, I think if we want a real answer/details on this, we shouldn't hash it out here, but instead email floatplane support and see what details they can/will provide.

 

I'll also say this, spoken audio doesn't need a much from a bit rate level as a full music bed. So this would be similar to the video comment I made above.

Edited August 18, 2023 by DoctaJones81
Adding audio note.