It looks like the new option won't work in the USB version until the option's text has been translated (i.e. 12.3, not a beta release). You'd have to switch Opus to English to see it there.
Ok.
Is the graph still considered largely unfinished at this stage? Will something be done to make it less worthless?
I'm glad I can now disable it, but I would much rather have a graph that serves a function and doesn't simply rehash the long-term average transfer speed. As it stands it really has no purpose and adds no value. Too many samples are computed in the average--we already have a long-term average in the text field above the graph. You can have an accurate current transfer speed without chaotic spikes in the graph by still averaging, but using far fewer samples. The graph was initially a big upgrade selling point for me and I was disappointed when I saw the implementation of it.
We'd still like to do some more work on it at some point. Since the graph is more about eye-candy to watch while waiting for copies than something that could ever be useful, and since making it nicer isn't a quick 5 minute task, other work has taken priority so far.
You've said what you don't like about it, but what would you like to see it changed to?
My feeling is it is a bit spartan at the moment and could do with maybe grid lines and a moving 'current speed' line across the graph, with a number on it. If you have other/different requests, let us know and we can consider them when we get to doing this.
I'd like to see numbers off to the side that scale depending on highs and lows on the graph. Like I mentioned, I'd also like to see the graph be more granular. It could still be smoothed (I don't mind a spiky graph, but I understand many find them unsettling) while not being smoothed to the point of merely reflecting the old average transfer speed text indicator above the graph. As it is, its indication seems redundant to me.
(Arbitrary numbers warning:) If, for example, the average transfer speed text indicator takes, say, 1 sample per second and averages them over the course of 10 seconds to give you the average transfer rate, make the graph indicator more precise where only the last 3 samples are averaged.
If you're suggesting a more precise current speed line overlaying the smoothed average we have now, I think that's a good idea as well. Only issue I see is that a dramatic dip or spike of the current transfer speed could send the line beyond the bounds of the graph defined by the high/low rates of the average transfer speed, or if the current transfer speed is allowed to define the range, then the range of the graph will be a little too wide to be intuitive or aesthetically pleasing until the dip/spike scrolls past the margin...
I understand it's not a priority and I'm happy with the option to turn it off for now. I'm glad to hear you're considering revisiting eventually.
Do you have an opinion regarding the other point I raised in the thread I linked previously?: This isn't related to the graph, but I think it makes sense to freeze the elapsed transfer time indicator when a transfer is paused. Maybe I'm misremembering, but I feel it may have worked this way at some point in the past... While the total active transfer time is an interesting metric for me, I don't see the use in knowing the total active + paused transfer time.
What would you do with that additional information? As in, would it help you make any important decisions about the copy operation that currently you are having trouble deciding on?
I see the only one reason for graph - if it works proper (scale down/up whole graph, not only new parts) - it can shows is file/disk fragmented (if some files copying with high speed and others are below, for example, 30 MB/s). And should show value while mouse is over (like in MSI Afterburner, for example). But IMO it's waste of time to improve this graph now. It's just visual firework and can be improved some day I think.
I would do the same thing I do with the trip odometer in my car, or the CPU fan RPM graph on my desktop--not much, but it's the kind of information I like to have instead of a graphical but wholly redundant rehash of the average speed text indicator just above the graph in the transfer window.
I'm a Directory Opus user--that alone should tell you I'm not the sort who only wants barebones functionality and information where it's only of mission-critical importance. Directory Opus has more features than almost any user requires. Why did I bother typing my previous message, providing what I thought to be the sort of well-considered feedback I had been asked for, only to be asked, sarcastically, for justification? What could possibly justify the existence of the graph in its current form?
I shouldn't have bothered to contribute feedback. It won't happen again.
The thing is, if you say "X is useless, I want Y instead" then it implies Y is useful, and we want to try to understand why/how Y is (more) useful to see if it makes sense to spend time on it.
It seems like both X and Y have no real purpose, other than being something to look at while waiting for the operation to complete, but you feel strongly against one and in favour of the other. That's fine if it's just a personal preference (I somewhat agree with the preference myself, although I don't feel strongly either way, and a lot of people asked for the graph so they got the graph), but it sounded like there might be more to it as your wording suggested the extra information might be useful for something.
Not sure why you responded in place of jon, leo, since I'm of the impression that you know the graph is in a sorry state at present. (I'm also not sure why I'm back.) Seems jon took his sarcastic potshot and fled.
I don't understand your talk of X vs Y. This is a false dichotomy, unless you're zeroing in on one part of the criticism: that the cursor of the graph shouldn't be redundant with the text above. I've said much more than that, both in this thread and the one I linked before. I'm in favor of X + Y + Z... where X is the graph and the following variables are additional features building on the premature graph feature. Do I even need to argue in favor of the usefulness of the other features, like numbers on the side that change based on the bounds of the graph? That's an elementary feature of any graph.
I was excited about the graph when I read about it, and I still like what it could one day be. It motivated, in part, my upgrade to DOpus 12, because it was something I wanted for a long time. Then I saw what little information is provides is redundant and there aren't even numbers on the side. The graph has no real purpose beyond providing eye-candy. The graph would have a purpose if it provided new information. Is it trivia? In most situations perhaps, but isn't unique trivia preferable to redundant trivia?
An indicator closer to real-time (fewer samples averaged together) would tell me more about how my drives are handling, e.g., lots of small files vs a few big files. Is it of critical importance? No, but what good reason is there not to have a second variable representing a more current rate in addition to the long-term average of the text indicator just above it? I've disabled the graph and I don't really care if I ever have reason to enable it again at this point. I still bothered to provide some feedback on how it can be improved and it seems I'm not alone in thinking there is ample room for improvement.
If you don't want to invest the time in making the graph better, that's one thing, but I find it hard to believe you can't see that what peterb, myself, and even you, leo, suggested would indeed be an improvement on what is currently a meaningless gimmick which I doubt many of the feature-requesters are happy with. Contrary to what you seem to suggest, no one is saying having a graph in the transfer window is a bad idea--my file manager from the early '00s had one (no, I do not wish to return to it). Only the current barebones implementation of it is problematic. The graph is never going to be a make-or-break feature, but it could add some value. It doesn't have to be useless (your word). I would understand if you or jon would just admit that you don't feel like taking the time to revisit the primordial transfer graph anytime soon--just don't pretend it couldn't use some work to get it out of the alpha stage where it so clearly is as of 12.3.
My intent was not to be sarcastic, but as Leo said an attempt to try to work out exactly what it is people want and how it would be useful.
I don't really understand how you can say the graph shows redundant information, since the information it does show (change in transfer speed over time) was not shown before and is not shown in any other form except as a graph.
Galneon, maybe think about the way you are wording things.
I agree that what is shown right now is "worthless" to the extent that any kind of speed graph is just eye-candy and the information shown will almost never be useful for anything other than distracting the user while they wait for the copy to finish. The graph is just something to look at, and that's probably an inherent property of such a graph.
(Personally, the only reason I have the graph turned on at all, now that it can be turned off, is to remind myself that I'd like to make it snazzier at some point. If I was just using Opus, not developing it, I would personally turn it off to make the progress dialogs smaller, as I don't find a speed graph to be worth the space it takes up, no matter what it shows or how it shows it. But that's just me, and a lot of people like looking at the graph, which is also perfectly valid. For those people, I'd like to make the graph nicer for them, and constructive suggestions about how to do so are welcome.)
So, unless there is some potential utility to the graph which we aren't currently aware of, suggestions about what the graph should and should not do come down to subjective opinions about what information is interesting and what kind of display is nice to look at. Not so much utility or worth, but look and feel.
To clarify Jon's point about the graph showing information that isn't shown anywhere else: He means there is nowhere else in the dialog you can see historic data or data over a period of time. The graph uniquely shows that in the dialog.
Making the graph use a smaller averaging window, so it is more spiky and responsive to quick changes, makes sense to me. That seems like the main suggestion within what is now a long thread. Units on the graph might also be nice.
These aren't bad suggestions, and we might do them in the future.
On the other hand, these suggestions won't make the graph any more useful than it already is, but you seemed to be saying they would, which is why we both wondered if we had missed some detail of the argument, or if you had other changes in mind which we had not considered. It seems not and you were saying something was "useless, alpha, primordial" because it wasn't how you'd like it to be, not because how you'd like it to be would be more useful in some way. And that's fine, just be aware that there are better ways to state that kind of request.
If the criticism of how things are also applies to how you're suggesting they should be, it makes people wonder if they have understood what is being requested, and that in turn makes them ask questions to find out if there is more to it.
1 Like
It supposed to shows that, but it's not true. At least in some situations. I described one - my HDD was fragmented and I start copying some big files from it to another drive. One file was fragmented and copying speed was about 20-30MB/s. Opus shows it as high level at start probably assuming that speed can only slow down, but not speed up more than 20%. Second file was not fragmented and transfer speed up to 100MB/s. What I was see on graph? That speed is... slower, because opus change range of graph, but only for second file without scale down previously drawn data. So this graph shows that speed was slower while speed was in the fact much faster.
Rest is just personal preferences - how graph looks, is showing values or not, gives access to previous data (scrolling back) or not. But not scaling down whole graph when maximum speed rises is wrong imo. Even standard win8 graph scale whole graph if needed. And imo looks pretty nice.
Anyway - I don't think that fixing graph is most important now and should have high priority.
Every information you need for file-management and also to notice if a drive works not as expected is there: Actual file/min.-max.-speed, overall time and size left.
I would suggest to remove the graph completely, as the requested things will end up in a complete and complex analyzer-graph - and IMO that belongs to "disk tools", not "file manager"!
Hi all. I don't mind the graph but is there any way to change the colour from 'Atari ST GEM Green'? I'm doing a backup and I'm sick of looking at it :oP Cheers!
Disable it! Or do you think you won't get sick looking on a blue, red,... one after a while?
Yeah, I've turned it off 
The colours come from the Windows theme. Progress bars use green for working, yellow for paused and red for error. The graphs are the same.
I like a speed graph, but those colors were the reason it's one of the first things i turned off. 
It's just too much saturation!
Maybe it can take inspiration from windows Task Manager instead? .. those graphs look very nice.
Devil's in the details. 