About heavy armor, and roundtime

[Skah]

Every other MUD might be pushing it a bit. Discworld comes to mind as one with a lot of command delays, and I've certainly played others which are comparable to Clok. Personally, I'd much (much) rather the GMs focus on their bug/feature lists (yay alchemy) than try to shave some time off of command processing. I'm not sure a big MUD written in single threaded Python with flatfiles for storage will ever feel as peppy as, say, a well-written LPC MUD.

The combat system rarely requires tons of commands in short periods of time, with the roundtime system, and I really don't notice the lag in feedback anymore.

[Lassyn]

I think it's reasonably fair to say that the lag has gone noticeably down from when I started playing over a year ago; I believe that Jirato did make a few of those fixes that he was mentioning.
But, as people already mentioned, I feel like a huge game with a lot of things to do can overcome some slight lag; if people are unwilling to figure out the lag (which to me is comparable to a huge unfamiliar world or an unfamiliar combat system as far as being something to adjust to) and they go to another mud because that is their deciding factor, then we wish them well and hope they'll come back eventually.
For now, I applaud the GM team (yay me) for working to bring new plots, activities, events, organizations, and features into what was, at its core, a learn-to-code project.

[Jirato]

Efficiency and optimization is always at the forefront of my mind when working with the sever. There are dozens, if not hundreds, of things to improve on. I know this, I recognize many of them, and am constantly making little tweaks here and there for the sake of performance. It's on ongoing battle that will take considerable time and effort.

Please do remember, I didn't write the server. I am maintaining it, adding to it, and slowly improving/optimizing it, but there are seriously over 100,000 lines of code that I did not write with almost zero documentation. I am finding new things every day.

Will CLOK ever get better? Yes. Slowly, but surely. The amount of rooms and critters CLOK supports now compared to how many it supported 4 years ago is a good sign of that.

Is there one single solution that I can immediately fix to provide a significant instant performance gain? Probably not. We're dealing with hundreds of little things here and there. They'll all add up to a lot when I get them resolved, but it's not going to be some instant overnight thing.

[Edoras]

I'm not familiar with Discworld and its command delays, but if that's a coded feature I think it's different than Clok's server lag, which is most certainly unintentional and also a performance issue at its core. For example, when connecting to discworld and poking about on the login screen I get a median return time of 234 ms per command with pretty low variance, as opposed to Clok's 1400+ ms median processing time that ranges anywhere from a virtually instant 30 ms return time to over 2 seconds in many cases. I realize that many players don't care too much about that sort of thing, but I also think that there are also many players, including potential new players, who notice the unresponsiveness of Clok's engine and it bothers them.

I also agree that all GMs should be focusing on what their strengths are. I wouldn't dream of suggesting that staff members who are doing a great job of enriching the game through plots and activities to be overly burdened with coding: However, for the staff members who are "in the code" as it were, I do think that this is something worth investigating so that at least the reason for the delay could be found, and I get the feeling that reason isn't known.

To put it differently, I can foresee four scenarios that could arise with regard to the server lag.

1. The worst case is that the devs spend their time working on things that don't fix the server lag, both taking time away from fruitful tasks while also frustrating them and making it less likely that the real reason will be found and/or fixed.
2. A better scenario is that the devs figure out what the real cause of the delay is, but then realize that the work required to fix it is too vast, and so they shelve the idea. While this still doesn't help anything immediately, it's technically better than case 1 because it should take less time to figure out what the cause is than to implement a fix that doesn't help much, and it leaves open the option of addressing it later if time allows.
3. The default case scenario is that we simply ignore it. No coding time lost, yet nothing gained.
4. The best case is that the devs figure out what really is causing the lag and are able to fix it, making the game very responsive and easier for new players to adjust to, while also encouraging existing players.

Right now I'm afraid we're in scenario 1: Where it's likely that the I/O of reading commands wasn't actually the source of the delay, or at least not a primary source. I'd at -least- like to help move away from that direction if that is the case, which is why I provided the evidence of testing the "time" command versus movement commands to demonstrate that the server delay is identical across them. I might be totally reading that wrong, however, and maybe there has been a performance gain as a result of those commands being optimized which I'm being too narrow-minded to see given that I just returned: As a side-note to that end, the game has "felt" more responsive than when I played before, although I don't have the numbers to compare to when I tried it out years ago.

All that said, while obviously it would be great if there was a single feature or single aspect that could be changed to instantly "fix" the server lag, it's just as likely that there are all sorts of reasons why the server lag is present, and it will only be resolved over many code changes for a long timeframe, as Jirato mentioned. I do at least hold the opinion personally that it's certainly worth investigating and pursuing within reason.

As a final note, Jirato, thank you very much for the work you do and for responding here. At minimum I felt like this was an elephant in the room, and I appreciate your responses. Just knowing that it's known by the staff and is being looked into is all I could reasonably ask for anyway.

[Acarin]

I personally enjoy a little bit of delay. Whenever I use the assassinate command, that second before I see someone drop is pretty exciting.

[Fayne]

Personally I think the minor delay is the least threatening problem to CLOK's player base. It's noticeable for some when they first join, but generally they all tend to stick around and see if the delay is worth dealing with (it definitely is). I think the biggest dangers to lost players lies in other areas, such as optimization of mechanics, implementation of new features, and new, exciting storylines and events of all size and flavor. All of these things would benefit more from valuable GM attention than the minor delay.