Viewer point of view: LTT videos

[Revonlieke]

Have not really been posting here much, but the latest honest answers video kind of made me decide to write this.

To me LTT has been the main source of general information about anything tech. Not disregarding other tech-channel, but I do think LTT has gone far from the old days.

But the topic I want to talk about is the confusion that a viewer experiences whilst watching the videos about tech-related stuff, more specificly;
videos about a new CPU/GPU or other PC component and posting results or comparisons between said components.

Now I am glad that bitwit actually went ahead and explained the 1% and 0.1% benchmarking. ( LINK )
That's actually really nice general information if someone decides to change the way they bechmark something.
But, it doesn't fix anything about a video such a this from being a confusing cluster of madness.

For me, from a subjective point of view LTT is really suffering from this. Whilst I do love the hard work the group put into each video,
I do think many of them could be divided into parts and try and focus on explaining things in a simpler way. Not saying I'm a toddler who doesn't understand a basics of an excel graph, but...

A good presentation of a graphics card and how it performs in a handful of games, to me is much more important, than seeing 5-10 other video-cards compared to it with a massive excel layout that doesn't make sense.
It is true that you still need to compare performance of something to something else, but adding more "comparison points" to one video grays out or even confuses to viewer a lot more.

For example; I never pause the video during the excel presentations. For one, they go by too fast for me to even get my hand on the spacebar and two, as I said; 
there's always 5-10 other CPU's or GPU's or whatnot being compared to the main one in the video and finding said component in the list of even 5 is hard, especially if it doesn't sit at the very top.

So instead of stuffing all the data you gather into one video. I'd hope LTT would cut the video's into parts like; Gaming benchmark, Video-editing benchmark, Zip-archiving etc... (though the last one might be a bit sort of an video)
And really just focus on what they are showing on the video.

Another example; "The Asus Hero GTX1080Ti sits right below the GTX Titan XP with, oddly enough; 2FPS defference with some margin of error..."
whilst some random diagram goes by in the video, doesn't help me at all as a viewer.

Instead, if your script talks about GTX1080Ti, highlight that card in the excel graph and when your script starts to talk about the GTX Titan XP, highlight that one.

That's just a very basic idea of how to improve the presentation, especially when you have a larger list of said components to compare to each other.

Another example is when people have started to compare testing rigs of AMD vs Intel. Listing ALL the freaking components in the systems.
Sure you want the viewer know that these are the same systems overall, but saying all the names in the video is boring to listen to, but also just listing names doesn't prove anything about it being legit or not.

But that's a hard one to fix, since I understand people want to atleast see the components in the rigs and see them listed as a WALL OF TEXT on the video.

I hope my thoughts about this were clear and I totally understand the "snappier" approach Linus has, but this is just what I've felt recently with especially the release of Ryzen.

I think Linus even said it himself that product names/titles are absurdly confusing as it is, so having all of those names/titles in one video doesn't really help all that much.

[Revonlieke]

9 minutes ago, FriendlyNonMurderingSort said:

Image

Hmm, I didn't color the text in any way though...
Seems to me like it's a mismatch with whatever layout you use.

EDIT: Fixed...

[Froody129]

The only relevant thing here for me is the time the graphs are in the video, I'm okay with pausing it but at least make it possible to easily pause at the right time. The rest was a nonsensical rant 

[Ezzy-525]

I find that either Linus or Luke tend to summarise the graph as they show it, so you don't really have to look at it anyway.

 

And you want the video split and to go into more depth, but you find it boring when they list specs?

 

Sounds like nitpicking to me buddy.

[Revonlieke]

As I said, it's mostly all from a subjective point of view.
It's just for me the summaries on top of graphs are very hard to follow, when you have ten different comparison points.
On top of it you have all the naming scheme's etc that make it even more confusing.

I do try to keep myself on top of all the latest tech etc.
But for sure when your summary consists of 7700, 7700k, 8350, 1800X and 6900 and what not it really just goes in from the other ear and out the other.

The forum in a sense is maybe a bad place to question this, since majority of you are very interested into everything LTT pushes out. What I'm talking about is the "avarage" viewer who IS interested in the video, but wants the bechmarks etc. to not confuse him even more.

Whilst I know that LESS information and "being simple" isn't actually good in cases where it is necessary to know for sure where some component sits in comparison.

But I do think there are some steps that could be made in the video-editing part and the script to not make it sound like you're being bombarded with all this information.

Again, it's subjective and partial reason might be that my first language isn't English, so it might not necessarily be as easy as cutting bread for me to listen and actually understand.

[Ertman]

I guess a good point maybe, is there has to be a way to simplify the information being presented, even the graphs. I don't necessarily mean dumb it down, but remove the clutter and simplify some of the presentation. Generally if I am watching a video, lets say a gpu review, I just want the information to get a basic understanding of how this card performs.

 

I think they should always post a link for an even more detailed breakdown for those that are interested.

[HelplmChoking]

9 minutes ago, Ertman said:

I guess a good point maybe, is there has to be a way to simplify the information being presented, even the graphs. I don't necessarily mean dumb it down, but remove the clutter and simplify some of the presentation. Generally if I am watching a video, lets say a gpu review, I just want the information to get a basic understanding of how this card performs.

 

I think they should always post a link for an even more detailed breakdown for those that are interested.

 

 

That's a good idea actually. I think the way the results are presented in the video is fine, but it would be nice to see the graphs included in the forum post for the video as well, just so we can refer to it more easily.

 

@nicklmg is that something that would be possible? Just adding in each graph as an image? Hopefully it wouldn't be too much extra work for you guys.

[BuckGup]

Wait...who asked you? 

[Princess Luna]

9 hours ago, Ezzy-525 said:

I find that either Linus or Luke tend to summarise the graph as they show it, so you don't really have to look at it anyway.

 

And you want the video split and to go into more depth, but you find it boring when they list specs?

 

Sounds like nitpicking to me buddy.

Not every one who watches them speak english fluently, aside the accent differences, being able to give yourself a look at the graph properly would help to understand what they are saying to a lot of people.

[BlueChinchillaEatingDorito]

Time to reset the "Days since someone told Linus how to do his job" counter.

[Princess Luna]

1 minute ago, BlueChinchillaEatingDorito said:

Time to reset the "Days since someone told Linus how to do his job" counter.

Why do we even have a counter any ways?

[BlueChinchillaEatingDorito]

Just now, Princess Cadence said:

Why do we even have a counter any ways?

Because it's such a common occurrence.

[Princess Luna]

Just now, BlueChinchillaEatingDorito said:

Because it's such a common occurrence.

So common we can throw it away, it'll never get pass day 1 any ways, get it? [:

[Yoinkerman]

Yeah a lot of the time they don't do a very good job of vocalizing the results of the graphs.  This has been brought up in the past, changed for a little bit, then went away again.  As a result I've mostly stopped watching their reviews because I either watch them at a low bitrate because I'm on mobile data, or am doing something else while watching the video (work or whatever) and can't look at the screen.

 

Sure; talking about the results is a little boring but what's on the slide should match the talking points, even if the talking points are expanded.

 

@nicklmg

[2FA]

They don't even do consistent benchmarking. LTT has stated they are more focused on the entertainment aspect and not so much the informative part.

[Wolther]

Just now, DeadEyePsycho said:

They don't even do consistent benchmarking. LTT has stated they are more focused on the entertainment aspect and not so much the informative part.

People watch tech videos for entertainment? 

[2FA]

Just now, Wolther said:

People watch tech videos for entertainment? 

Apparently so according to Linus.

[RedSpade]

I 100% agree as recently the graphs are quite hard to grasp in the time without breaking the immersion with pausing. I think the best option for LTT is to order the tested tech by cost with most expensive at the top. It gets very confusing when the order of the graph constantly changes and you have to find what cpu/gpu you're comparing in every single slide and then repeat the process again 3sec later.

14 hours ago, Revonlieke said:

Another example; "The Asus Hero GTX1080Ti sits right below the GTX Titan XP with, oddly enough; 2FPS defference with some margin of error..."
whilst some random diagram goes by in the video, doesn't help me at all as a viewer.

Instead, if your script talks about GTX1080Ti, highlight that card in the excel graph and when your script starts to talk about the GTX Titan XP, highlight that one.

That's just a very basic idea of how to improve the presentation, especially when you have a larger list of said components to compare to each other.

Something like this as well I believe would be super helpful.

 

Another idea that may be helpful to find out what people actually like is to run a survey on Floatplane Club where you upload a few videos that are talking about the same thing but the presentation of the graphs changes. E.g. video A has it exactly the same, video B implements my suggestion of ordering my cost of cpu/gpu, video C has @Revonlieke suggestion of highlighting what you're talking about and video D is a mix of B and C. Then you will find out exactly what we all want, with hopefully not as much work on your part.

 

@LinusTech, @Slick would something like this be possible?

[mrchow19910319]

If you want more informative data, go to hardware unboxed...

LTT's video is more like a ad that has some "entertainment value"....

They've been staying away from the "informative" path long time ago. 

 

And to be honest, I dont really enjoy it. It makes me feel that they dont take their videos serious enough. 

I am pretty sure that they did do some benchmark, but way too little to give you the big picture. 

 

[LinusTech]

The reason the benchmarks are blowing past at superspeed is that we want to put them in there to show "hey we actually did the homework" but we don't honestly think the exact difference in scores between one piece of hardware and another is necessarily that important.

 

We've talked about it internally and I think the direction we're going to head is a summary voice over of the key conclusions we were able to draw from our testing based on the numbers we collected.

 

The raw information will be there for folks who care enough, and the bigger picture stuff will be summarized - all without increasing the length of our videos.

 

Information density is something I personally value as a consumer of content and is something I continue to strive to increase in our own videos. It's not necessarily the best for non-native speakers, but it's been one of the things that I think has led to higher quality so far, so I'm not looking to reverse the direction.. yet..

 

As always, making the perfect tech video is a moving target...

[LogicalDrm]

Ignoring post above.

 

Main problem is that there is too much information crammed in under 11 minutes. Or whopping 13mins of Ryzen 7. Which in general is mainly issue of targeted audience. Shorter video, more chances of someone less interested of watching it until the sponsor spot at the end. Graphs are there just to give some hint of whats going on. Main issue is that they are really hard to read. With too little options and too much color. Like the reviewed item(s) aren't highlighted. But rather its brand highlighting.

[Yoinkerman]

5 hours ago, LinusTech said:

The reason the benchmarks are blowing past at superspeed is that we want to put them in there to show "hey we actually did the homework" but we don't honestly think the exact difference in scores between one piece of hardware and another is necessarily that important.

 

We've talked about it internally and I think the direction we're going to head is a summary voice over of the key conclusions we were able to draw from our testing based on the numbers we collected.

 

The raw information will be there for folks who care enough, and the bigger picture stuff will be summarized - all without increasing the length of our videos.

 

Information density is something I personally value as a consumer of content and is something I continue to strive to increase in our own videos. It's not necessarily the best for non-native speakers, but it's been one of the things that I think has led to higher quality so far, so I'm not looking to reverse the direction.. yet..

 

As always, making the perfect tech video is a moving target...

Personally I'm just asking people who both watch and listen get roughly the same info.  For non benchmark videos it's not an issue at all and that's great.

[NuclearKing]

6 hours ago, LinusTech said:

The reason the benchmarks are blowing past at superspeed is that we want to put them in there to show "hey we actually did the homework" but we don't honestly think the exact difference in scores between one piece of hardware and another is necessarily that important.

 

We've talked about it internally and I think the direction we're going to head is a summary voice over of the key conclusions we were able to draw from our testing based on the numbers we collected.

 

The raw information will be there for folks who care enough, and the bigger picture stuff will be summarized - all without increasing the length of our videos.

 

Information density is something I personally value as a consumer of content and is something I continue to strive to increase in our own videos. It's not necessarily the best for non-native speakers, but it's been one of the things that I think has led to higher quality so far, so I'm not looking to reverse the direction.. yet..

 

As always, making the perfect tech video is a moving target...

Keep the benchmarks, I just pause when I want to look at them in more detail. The extra density of information has been greatly appreciated!

[Sniperfox47]

I'm totally on board with Linus' explanation here. The only real use of the graphics benchmarks is so that you can make sure the data is actually saying the same thing they are.

 

I mean if you're one of the 20 people who play Ashes, like me, then an Ashes benchmark may be useful to you, but overall it's the summary that's the useful part.

 

If you look at the results, typically they're set up to show it against a bunch of similar competition in different ways so you can better judge it's value for your use case. Like for the massive 18.4" MSI GT83 Titan SLI, it was compared against:

 

a comparable desktop with and without SLI - after all the Titan *is* a desktop replacement, and does use SLI, but a non-sli desktop might be better since SLI has some serious issues.

The Acer 21X which is much bigger at 1.5x the weight and offers much the same performance.

The Sager NP9873 which is a smaller 17" beast

And a XPS 15 which shows you a much more portable option

[Revonlieke]

Well good the hear there's actually some actual merit to what I'm saying. Clearly there's  mixed feeling about this.
 

11 hours ago, LinusTech said:

The reason the benchmarks are blowing past at superspeed is that we want to put them in there to show "hey we actually did the homework"

 

The raw information will be there for folks who care enough, and the bigger picture stuff will be summarized - all without increasing the length of our videos.

These are pretty much the things that's I am indeed talking about. Whilst I do understand your take on it, here's my immediate thought on it;

Both "we did our homework" AND the raw information are targeted towards the people who do care enough, like you said.
I'm not saying I'd like to "consume" false information or that I don't care, what I am saying is that some of the homework could be set aside into a forum post or something, instead of it all cluttering a video.

That being said, I did like the presentation Does RAM speed REALLY matter? had. It was closely perfect in terms of presentation, video editing and talking points. With some minor confusing parts. Example at: 4:12, comparing the two systems. I was immidietely thrown off by the comparison of the two systems in regards of what ram speed they ran at. It lists the two systems side by side, but I had to find RAM-line and since it was listed as a "spec list" I wasn't sure if you just used 3200MHz Ram or, like you said, achieved 3200MHz.

In this small example you could have just listed the memory sticks used with their max. potential MHz and underneath that, have the "achieved" version.

All the rest about the specs of these systems was redundant information and clutters the point you're trying to make and makes it confusing.
But I'd totally also understand that there's be an angry mob running around speculating those two systems we not the same