Opinion on OpenComputers?
#1
Posted 19 December 2013 - 08:52 AM
If you want to post a status that 'hates' on the mod, please don't make it inappropriate.
#2
Posted 19 December 2013 - 10:05 AM
There are like 5 suggestions a day I like in the CC forum, but the last suggestion accepted were colored turtles (which you don't need at all, I use turtle's to mine and not to play, and most CC Games have a fixed solution of 51x19, which turtle's don't support), and it seems OpenComputers opens much more opportunities. Also, I don't like this Low-Tier in ComputerCraft.
(I'm not a hater, I still like ComputerCraft much, but OpenComputers seems more attractive ^^)
The only reason that speaks against OpenComputers is that it uses lua 5.2 and setfenv doesn't work anymore (which is like inacceptable) (but maybe you can make Computers that use lua 5.1)
#3
Posted 19 December 2013 - 10:11 AM
ComputerCraft is low-tier, but probably as a way to introduce you to programming
Would you object to using them together?
#4
Posted 19 December 2013 - 10:14 AM
#5
Posted 19 December 2013 - 10:59 AM
It will be interesting to have a competing mod out there. In a couple of threads over on reddit, it was mentioned that dan200 was actually in the middle of a big chunk of work on ComputerCraft. He also mentioned reconsidering his decision to put off adding persistence. I'm hopeful that the presence of a competing mod will cause both mods to become better than either would have been alone. I'm looking forward to seeing the changes dan is working on!
I haven't looked into OpenComputers very deeply yet. I know it uses Lua 5.2, which I'm somewhat hoping the new changes to ComputerCraft will avoid moving to. I'm somewhat doubtful that programs written for ComputerCraft will be cross-compatible. If there are enough similarities, I'm sure it will be possible for someone to come up with a compatibility layer, of course. The modularity concept is an interesting one, though it appears as if they've gone the multiblock route, severely inhibiting computers' roles in space-restricted builds. I hope that ComputerCraft will continue to provide compact and powerful solutions.
That said, a word about OC and Ask a Pro: these are the ComputerCraft forums, so questions in Ask a Pro should be ComputerCraft-specific. We haven't encouraged general Lua questions, nor questions for other games using Lua. This holds for OpenComputers. We won't try to prevent questions being asked unless it becomes a problem, and we'll address that on a user by user basis.
#6
Posted 19 December 2013 - 11:29 AM
EDIT: found it.
Looks like a pain to install... CC FTW!
Edited by Castform, 19 December 2013 - 12:49 PM.
#7
Posted 19 December 2013 - 05:54 PM
EDIT: Just now remembered it needs to be powered, but that's not that big of a deal
Edited by Death, 19 December 2013 - 05:55 PM.
#8
Posted 19 December 2013 - 06:21 PM
Edited by oeed, 19 December 2013 - 09:41 PM.
#9
Posted 19 December 2013 - 06:35 PM
- OpenComputers(OC) is open source. I like open source.
- OC looks like it's running the C implementation of Lua, hence having to install all those libraries, and Windows/Linux only. Winner here is ComputerCraft(CC) for picking LuaJ, simplifying installation.
- OC computers are really expensive and require more than one block or item (case, keyboard, screen, gpu, memory, power converter)
- OC computers also need power from somewhere
There's a real lack of innovation in ComputerCraft and I feel that it's partly due to this idea that ComputerCraft is complete, and that anything else you could possibly want to do you can by writing the Lua yourself. The problem is that leads to a wasteland of APIs and OS where everyone has to reinvent the wheel in their program because the base system doesn't have it built in.
I hope OpenComputers causes innovation to occur in ComputerCraft and it's community.
#10
Posted 19 December 2013 - 06:46 PM
I'd like to see the mods remain different enough that they'd both be included in the main modpacks out there. While I don't think many people care how much a CC computer costs, CC turtles are incredibly cheap for what they do and a more "balanced" replacement could be their downfall. CC's stuff is only low-tier in terms of resources required. OC's stuff costs more and gives you less, but seems to have about the same potential.
(Hands up if you only found out about CC because it was included in a modpack?)
Rather, I'd like to see OC perform some tasks CC can't, and CC perform some tasks OC can't. It'd potentially make for some much more interesting builds, too, given that the two different mods have the potential to communicate (over redstone at the very least).
What I would not like to see is the sort of power creep that starts to occur when mods start to "compete". Fortunately, I don't think that's going to happen here. Sure, some of the rejected suggestions for CC will likely make it into OC, but that's got a lot more to do with them being "suitable" for OC than it has to do with them being "great, original ideas". On that note, it's very seldom that I see a "suggestion" made on these boards that isn't 1) obviously a bad idea, 2) repeated all the way down the page of threads in the suggestions board already and 3) flagged as a "don't suggest this" point in the stickies. Usually all three of these apply.
On the subject of persistence, I've always rather liked that CC computers lose their state, even if I must admit that other than the technical ones I can see no good reasons for them to do so from a player's perspective. I feel it encourages better code (or at least, encourages people to think) - users need to work out how to make their systems handle random "power outages". I realise that most methods that actually get used are waaaay more inefficient than they should be (eg this one, which I quite liked, has a crazy amount of overhead even compared to... well... any other I've seen), but still. It's more than possible to code a moving turtle that can boot up from any location and get on with its work without any filesystem access at all, and quite satisfying to do so.
I'm trying to think of a downside to persistence, and I rather suspect the main one is that a bad enough server crash will probably foil it anyway (meaning that users may STILL need to make their code deal with random "power outages", but those functions will just get called a lot less often). That and the "OS X users can only use the mod by joining a server that runs it" thing his implementation enforces, but I've never understood OS X users anyway. So yay for persistence, I guess.
Edit:
distantcam, on 19 December 2013 - 06:35 PM, said:
Edited by Bomb Bloke, 19 December 2013 - 06:49 PM.
#11
Posted 19 December 2013 - 07:01 PM
Bomb Bloke, on 19 December 2013 - 06:46 PM, said:
#12
Posted 19 December 2013 - 07:40 PM
My impression is that CC is largely intended to let you do what you want while making sure you understand how you're doing it, and additional "convenience" APIs take it in the opposite direction - assuming I'm right about all that, you can see why related "suggestions" are shot down time and time again. Maybe OC will be different in that regard. I don't know, but I really hope it's not. If anything, it sounds like configuring a system for OC will be even more complex then in CC.
As for the repeated operating systems? I've not ever seen the need to use an alternate OS on a CC system - the base shell does all I want. Even assuming I'm easily pleased, I don't think there's anything anyone could add to it that'd stop the flow of new "OSes" being written. And as long as people find them fun I don't see anything wrong with that.
#13
Posted 19 December 2013 - 08:39 PM
Bomb Bloke, on 19 December 2013 - 07:40 PM, said:
Anyways, on topic now, I'm really excited to see this mod. I'm not particularly interested in using it, but purely the fact that it provides some tough competition to the CC Devs (dat Unicode support) has me eagerly awaiting how this friendly rivalry will play out. No doubt, the benefits will be bidirectional.
#14
Posted 19 December 2013 - 08:49 PM
I'll agree that a lot of suggestions are bad, and the really bad ones are worth shutting down. But sometimes a discussion can arise that will turn it into a good suggestion. Or like this thread where some interesting ideas have been raised.
Enhancing the base OS won't stop the flow of OS, but it also won't hinder it either. If someone wants to write their own OS they can. It's the same with real computers, people are free to write their own kernel from scratch, or use an existing kernel and build their own OS around that, etc etc.
A good example of an unloved API that is impossible to implement in Lua would be the HTTP API. There are a lot of websites with APIs nowadays and most of them are inaccessible to ComputerCraft because they require OAuth authentication, which is impossible to do with the current HTTP API.
Back on topic, I hope OC helps push CC forward, and vice versa.
#15
Posted 19 December 2013 - 09:28 PM
distantcam, on 19 December 2013 - 08:49 PM, said:
Yes, yes, yes, yes, YES! I have wanted support for being able to open browser windows in the real internet browser (standard Minecraft link prompt) and support for headers, etc in requests to websites to make various things, such as a GitHub GUI (as well as a modified version of Git in the terminal) in awsmazingOS, and more. I would also like some support for parsing JSON returned by websites, because right now I have been writing my own handlers.
#16
Posted 20 December 2013 - 01:28 PM
#17
Posted 20 December 2013 - 04:19 PM
#18
Posted 20 December 2013 - 04:28 PM
I don't like the power requirement though.
And it crashes my game when I place it's blocks.
#19
Posted 20 December 2013 - 04:51 PM
Lyqyd, on 19 December 2013 - 10:59 AM, said:
gollark8, on 20 December 2013 - 04:28 PM, said:
Edited by Sangar, 20 December 2013 - 04:52 PM.
#20
Posted 20 December 2013 - 04:57 PM
gollark8, on 20 December 2013 - 04:28 PM, said:
I don't like the power requirement though.
And it crashes my game when I place it's blocks.
EDIT: Damn ninja
Edited by awsmazinggenius, 20 December 2013 - 04:58 PM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users