298 replies to this topic

[Li Song]

LSML 1.6.4 Jumpy Jets
This is a maintenance/bugfix release to 1.6.3

Found a BUG? Have a great idea?
Please report any bugs/issues/feature requests here: https://github.com/E...eporting-Issues. Please do take the time to register and post, I can't fix what I don't know about.

UNINSTALL THE BETA FIRST!
There was a slight mishap with the version numbers on the beta releases, meaning that 1.6.4 will not install until 1.6.3-betaX is uninstalled manually.

Download from the main site: http://li-soft.org

Changes since 1.6.3
* Jump jets that were fixed in omnipods are correctly handled.
* Seconds field of ammo table now correctly updates when modules are added.
* New and much improved handling of quirks. All mech quirks are applied to your stats.
* Loyalty mechs were incorrectly identified as champions and hidden by variant filter.
* Importing clan mechs from smurfy garage would fail.
* Command console can now only be equipped on eligible mechs.
* Updated parsing.

[J-Pax7]

I tried to install the program for the first time yesterday and got the following error when starting the program

'unable to start receiver java.lang.nullpointerexception'

It then shuts down. Is this a known issue or?

[Li Song]

@Paul Crux
I've never seen that before. Which java version are you using?

[J-Pax7]

Well if I checked the right place its 8.0.250. I t should be the most recent as I updated it after I got that error.

[Nori Silverrage]

Thank you for this. I love it and use it all the time. Seeing the actual values after quirks and modules is amazing and I don't need to be online or logged in so that is nice as well.

Have come across a bug though. Heat containment doesn't apply properly. It just adds 0.1 (as opposed to multiplying by 1.1) no matter what I try.

[Li Song]

@Nori please use the github issue tracker for reporting bugs. Instructions are given on the download page and the main site. I check this forum rather rarely.

LSML 1.6.5 Bullying Bugs
This is a maintenance/bugfix release to 1.6.4

Found a BUG? Have a great idea?
Please report any bugs/issues/feature requests here: https://github.com/E...eporting-Issues. Please do take the time to register and post, I can't fix what I don't know about.

UNINSTALL THE BETA FIRST
There was a slight mishap with the version numbers on the beta releases, meaning that 1.6.4 and above will not install until 1.6.3-betaX is uninstalled manually.

Download from the main site: http://li-soft.org

Changes since 1.6.4
* Smurfy export SSL error fixed.
* Equipment positions now stay the same after saving and opening the garage and importing loadouts from smurfy.
* A Max DPS graph has been added in addition to the max sustained DPS graph.
* Clan CT is now visible and editable in compact mode.
* Heat Capacity and Time to Overheat values now calculate correctly with coolrun/heatcontainment.

Edited by Li Song, 22 February 2015 - 08:44 AM.

[Nori Silverrage]

Will do in the future. Thanks for the release.

[Knight Lightning]

Li Song, on 17 September 2013 - 12:43 PM, said:

Introduction
Li Song Mechlab (LSML) is an EFT-inspired, open source tool for configuring and analyzing loadouts for PGI's MechWarrior Online.

The target audience for LSML is intermediate to advanced pilots who already have a rudimentary understanding of the game mechanics.

In Li Song Mechlab you can configure and share your 'mechs, try different loadouts, see how many tons of ammo you can fit and see how many seconds of continuous fire that will give you, balance your weapons for most efficient heat management with damage graphs and much more.

Hello Li Song,

Are you not worried about the Java security exploits, when even Chromium are going to make it impossible for Chrome to support NPAPI plugins in their browser soon? (You can run NPAPI plugins until Chrome version 45) And, according to data from the 2014 IBM X-Force Threat Intelligence Quarterly Report, the favourite target of security exploits is now Java.

Does this not bother you? I mean the program looks great and I have DL'd it, though I'll only use it on my notebook, that is not connected to the 'net.

Regards,

Knight-Lightning.

[Li Song]

@Knight Lightning
In short, no not really.

I have been careful in programming LSML, and specially when dealing with user input. LSML only ever connects to the internet if you ask it to download your smurfy mechbay. And even then it's using the Java default library for HTTP transfer.

The above facts means that LSML is an unlikely infection vector in it self. If I get reports of a security vulnerability in LSML I will patch it promptly.

As for exploits targeting the JVM, that is out of my hands. You should use a firewall and keep your JVM up to date. Possibly use another JVM. At any rate there is nothing I can do about this.

Edited by Li Song, 10 May 2015 - 08:48 AM.

[LMP]

I was playing around with LSML and noticed that it said the C-LRM 5 does more DPS than the C-LRM 20.



Is this a bug in the software or is it true that the C-LRM 5 does more DPS than the C-LRM 20?

[Gneckes]

Iirc (I'm at uni and can't check, sorry), the LRM 5 really does have better DPS than the LRM20, due to much lower reload time.

[Li Song]

@LMP

I'm not sure what you're seeing in the graph but to me it shows that LRM 5 does less than LRM 20... I guess you have two LRM5s vs one LRM20 which is why they look about the same size.

It's a stacked graph btw.

Edited by Li Song, 12 May 2015 - 02:05 PM.

[LMP]

Li Song, on 12 May 2015 - 02:05 PM, said:

@LMP

I'm not sure what you're seeing in the graph but to me it shows that LRM 5 does less than LRM 20... I guess you have two LRM5s vs one LRM20 which is why they look about the same size.

It's a stacked graph btw.

Yeah that's exactly what I had two 5LRMs and one 20LRM.

I also want to say that this thing is really a great tool and has a very professional look and feel to it, I really like it. Great job Li Song!

[Li Song]

I've found my self with some time to work on LSML and gotten around to starting working on the long overdue and often requested "Weapon Lab" coming in 1.7.0.

Below is a sneak preview of what I have cooking right now, most of it is working already. However at this point I'm afraid that the right hand side is too cluttered and I'm asking for some input. Do you think it is cluttered and hard to read? Does any one have any suggestions for how to improve it?

Edited by Li Song, 06 June 2015 - 10:06 AM.

[Li Song]

Update from last version. Text is properly packed and aligned. All fields update and calculate correctly now. I have moved the tabs to be "inside" the mech view. So that the overview together with the check-boxes for the upgrades and efficiencies (on the right) are always visible.

I'd be very happy to hear some comments and feedback on this design idea.

[Quicksilver Aberration]

So, I have been trying to get custom mechs (like those yet to be released) finagled into your mechlab for the sake of quirks etc and I have a question. In your datacache XML file, does the id value have to be contiguous and/or does it have an upper bound or could I set it to be a arbitrary value to allow for more compatibility with patches that add/remove stuff?

I did manage to add a single Wolfhound variant, but realized there may be issues trying to copy and paste if an update adds or removes a single quirk or something like that.

[Li Song]

I'm using the XStream library with ID references for writing the datacache so the behaviour/requirements depends on their implementation. And I haven't tried to do that by hand so this is mostly speculative.

But to answer your questions:

1) The ID's have to be unique. I'm 100% sure on this.

2) I don't think they have to be contiguous. Try it and let me know what happens

3) I would assume that they use a 32bit integer for the ID so a safe upper bound is: 2^31 - 1 = 2147483647.

Hope it works out for you!

[Quicksilver Aberration]

Li Song, on 09 June 2015 - 11:36 PM, said:

2) I don't think they have to be contiguous. Try it and let me know what happens

Unfortunately it seems they have to be, I added the Wolfhound with contiguous IDs and then tried it with some spacing in between and it did not like the non-contiguous IDs.

Just to clarify in case I am using the wrong word to describe this:
ID sequence 1: 0,1,2,3,4,5,6,7,8,9 (Acceptable)
ID sequence 2: 0,1,2,6,5,3,4,7,8,9 (Acceptable)
ID sequence 3: 0,1,2,3,4,7,8,9 (Unacceptable)

Also, I may have something else going wrong, but this latest patch broke your parser.

[Li Song]

Hmm well I didn't expect that... unfortunately as it is a library, it is out of my hands. :/
I'm aware of the parsing, it's MASC they introduced, I haven't had time to look at it yet.

[Li Song]

I will try to have a patch out (1.6.6) on wednesday next week (1st July)