Science & Industry

Very occasionally, I've had the same thing. I think the way I reproduced it was when removing multiple blueprints in quick succession. I'll add some additional checks to try and fix it.

Being in Australia, you will have the worst experience I could expect. I'll do some more local performance testing and simulate a really bad connection and see if I can make things better.

Thanks for all the feedback so far!

Manufacturing tab, add two control towers, set ME/TE on both, set region and build station, click on teams, get message that item has already been added, crash. I'm learning to save my batches more frequently lol.

Also, the teams feature is too noisy. There should be a tickbox to show only teams in the current build system.

I've released a new version. All the groups are now present in the manufacture tab including control towers.

Another part of the update is automatic error reporting if you get a crash. That should allow me to track down why you're getting these strange problems.

With the latest update, any time I click on the manufacturing tab I get a lost connection message and can't continue. I checked again after an hour and it's still happening.

But overall I'm damned excited about the progress, as this is hands down the best profit research tool available.

Moved to support channel.

public double ProfitToChildRatio
{
get { return Margin/(manufac.TotalChildBuildTime + manufac.ManufactureTime); }
}

Take the margin and divide it by the time required to build all the components for that item and add that to the time required to build the final product. It was something I found myself wanting to know fairly regularly; it may not have much use for others. The higher the number, the more profit you're getting for time spent building components. Just because something has a higher margin doesn't mean it's necessarily going to be better to build when you take in to account the time to build the components as well.

You're probably right that for maximum volume, you need to be buying components and not building them. I'm happy to build them, even if it means a reduction in the PPH because I don't want to be spending all my time ferrying components back and forth. That said, one of the early testers for EMT started buying components to keep their production lines running. They soon found that the market couldn't support the amount of components they needed within the time they wanted them and so they're also building components.

A feature planned for future release is to allow you to indicate whether you want to buy some components (thereby removing their materials from the shopping list and using the price of the components for the cost, rather than the materials) or just leave them being built. Maybe I'll get that done sooner rather than later so we can compare the difference between the two.

Yea, I'm still trying to work out why it isn't loading for you. I can load it just fine locally so I expect the problem to be something obscure.

Once I get the "I'm buying this" checkbox and associated changes sorted, perhaps I'll reevaluate how I build things. For now, I'm quite content with waiting a few days and making more per item, even if it doesn't allow me to push a larger volume through.

1) Yes but with the change mentioned above, it'll take that in to account.
2) I suppose so, yes. The profit per hour column at the moment only applies to the final product though. For example, if I'm building 1 scimitar which takes 36 hours with a margin/profit of 36 million, that's a very simple 1 million profit per hour. If you're building components though, you may want to know how long you spend building everything versus how much profit that thing makes. I suppose I could rename the column something like "entire job pph".

You're right about the number not meaning anything. I've not given much thought as to what unit it represents but it's probably profit per second. If I multiplied it by 3600, it might give a more useful value.


Are the highlighted items what you have selected from the manufacture pane? I'm not sure I understand what you mean by "some indicator to show items where the profitability is impacted by selected teams". What is it you want to see for that?

Stability has improved with the latest update, possibly because timeouts settings have been increased. However, performance is still badly suffering. My latency to the server is consistently about 350ms, so there must be something chatty and synchronous happening in there somewhere. Trying to add the Fuel Block group spends literally minutes, as does changing the build system -- in the pricing area maybe?, and saving is also slow. Once loaded, the grid is generally responsive.The following tracert suggests the biggest cause is within your isp?

1 2 ms <1 ms <1 ms 192.168.0.1
2 22 ms 8 ms 19 ms 10.97.0.1
3 11 ms 18 ms 9 ms sun2-ge13-0-2.gw.optusnet.com.au [198.142.164.5]

4 25 ms 25 ms 28 ms 198.142.250.193
5 23 ms 23 ms 22 ms 198.142.139.114
6 27 ms 23 ms 23 ms 198.142.139.130
7 180 ms 186 ms 181 ms 203.208.190.193
8 194 ms 194 ms 193 ms te0-0-0-9.ccr21.sjc04.atlas.cogentco.com [154.54
.10.169]
9 195 ms 193 ms 192 ms be2016.ccr22.sfo01.atlas.cogentco.com [154.54.0.
177]
10 238 ms 229 ms 252 ms be2133.ccr22.mci01.atlas.cogentco.com [154.54.30
.66]
11 245 ms 242 ms 256 ms be2157.ccr42.ord01.atlas.cogentco.com [154.54.6.
118]
12 255 ms 255 ms 252 ms be2115.ccr22.jfk02.atlas.cogentco.com [154.54.6.
189]
13 338 ms 336 ms 398 ms be2490.ccr42.lon13.atlas.cogentco.com [154.54.42
.86]
14 328 ms 332 ms 330 ms be2036.rcr21.lhr01.atlas.cogentco.com [130.117.4
8.154]
15 411 ms 339 ms 376 ms 149.6.8.202
16 342 ms 349 ms 344 ms customer-46-227-148-209.gigaclear.net [46.227.14
8.209]

Would it make more sense to do grouping by blueprint groups rather than item groups? For instance Battleship Blueprint #107 instead of Battleship #27. There are quite a few items where the blueprint categorization works better than the item, and I don't see any cases where it's worse.

Edit: Back to unable to connect. Start the program, log in, do nothing, Lost connection message within about 15 seconds. Still 350ms latency so nothing is obviously worse about the path to the server. I have two accounts now, one empty, one populated with several batches. Both can't connect.

This release seems to have been a big step in improving latency issues. However, I try not to close it very often as loading the mfg tab takes about a minute -- which I think you've said you're working on. But, the only way to refresh team data is by reloading. Put an expiry on cached team data?

Overall, once the data is actually loaded, the usability is fantastic. The ability to drill into groups, and then X out the noise, is just inspired, and the main reason I prefer EMT to the competition.

Tweaks to consider depending on level of effort and overall value:

- Retain column settings: current sort and widths. Possibly apply current batch settings to all batches? This would be mitigated if I didn't have to reload to see current team data.
- Ability to rename a batch
- I'm learning to remember to set pricing region and build system before I load blueprints, but perhaps remember the last value used when creating a batch?
- Refresh pricing loads cached market data. When I look at an item in-game, EveMon uploads it and it shows in EveCentral immediately. Your server is using cached data, but any chance that Refresh Pricing would cause items with no pricing to hit EveCentral again? (Only if the performance hit isn't fatal of course)
- PE should be renamed to TE
- I continually forget to set ME/TE in blueprint defaults before loading a group. Surely 10/10 is better than 2/2 or 0/0 as a default :)

1) I already have a note to sort out a better way of reloading teams on the client and it's one of the more important things to include right now. I'll consider being able to save column widths/sorts but that could be a lot of effort for not much gain.

2) Already exists; not obvious. Double click on a tab header and you'll get a box to allow for renaming.

3) The intended workflow for EMT was to start with an invention batch and then manufacture. Perhaps a user option on the options tab to set a default pricing region/build system? That sounds like the best place for it.

4) I'm fairly sure that as long as the batch isn't locked (that freezes material prices), pressing refresh will cause a fresh set of data from eve-central to be pulled. I don't know if eve-c has any API caching. I'll check to make sure it works like that though.

5) *sigh* Yea.... effort! I'll put it on the list.

6) The defaults are mostly used for invention. Now I think about it, I can just hard-code the 2/2 as the default. Perhaps again, a default setting in the options tab if you haven't set a specific ME/TE?



7) Agree on the excluding T2/faction items. Where are you seeing meta though? I didn't think there were any blueprints for those.

8) This would be taking the volume of the item and doing something with the IPH to give a rough "true IPH" for that item? I'll have a look at this although I don't think the eve-c API reports volume terribly well.

9) (manufacture cost + install cost) / quantity. I agree though, tooltips to indicate where a value comes from could be useful.