-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
IncGrid.tbl has an extra line of data. What's that about?
Another kind of smoothing trick?
Another kind of smoothing trick?
Okay, my guess is that the extra line of data indeed indicates a mechanism to facilitate (P)TI smoothing. My guess is that each row/line of blocks in incgrid.tbl affects the lightmap blocks both above and below it. If this is true, then it is also likely that incgrid blocks affect both left and right lightmap blocks.WiSK said:IncGrid.tbl has an extra line of data. What's that about?
Another kind of smoothing trick?
So a rough guess at the algorithm. In any given lightmap block, we make a mask for each province. We check the four neighbouring incgrid blocks { (x, y), (x, y+1), (x+1, y), (x+1, y+1) } for a reference to that province, then we can determine the level of 'PTI fog' from four directions (Inferis calls this the 'anti-alias' value). Then we draw a smoothing block in a similar way to the lightmap smoothing using the four fog values, then we use the mask to cut this to the shape of the province within the lightmap block.
Am I on the right track? It's probably more complicated that this, looking at the screen in EU2...
No, that's incognita.tbl.tombom said:
I have just tried my algorithm and it gives a reasonable result when overlayed on the id.tbl map. I'll post a screenie later.
EDIT: On second thoughts, it's not such a great result. I'm starting to doubt that I'm going about this the right way.
Last edited:
Count the clock cycles yourself?Inferis said:I'd just like to say: what would one do without a code profiler?
EDIT: any comment about my incgrid musings...?
Last edited:
Not really. I'm not sure about this format myself. I know that the 4 lower values are ids (as specified in on my TBL pages), and the 4 other are weights. They define how foggy that part has to be drawn. I'm still looking at how the actual drawing is done...WiSK said:
Well, per block there are a few cases to consider:Inferis said:
A) nation knows all the provinces -> no fog
B) nation doesn't know any of the provinces -> maximum fog (?)
C) nation knows some provinces -> gradated fog, depending on neighbouring blocks.
Case C is the interesting one of course. Does the engine look at the same pattern of blocks as with the lightmaps (to the right, below and below-right)? Or does it look in all eight directions? From looking at the data, I would guess the former, but looking at the map, I hesitate.
Something that made me discard my first theory (seven posts earlier) is that the fog overlaps province boundaries.
You sure? How does that help me exactly? Along the border blocks, incgrid contains references to the neighbouring provinces, so I already kinda know the adjacencies which are relevant to that block.Inferis said:
Something I noticed around the Bering Straits/Aleutian Sea is that even though incgrid has not-max-fog values throughout the province, the game still renders that area as max fog. This could be down to the colorscale used (ie any fog value lower than 10 is effectively max fog) or it could just be that I was correct about case B: if the player/nation knows none of the provinces, then max fog is applied.
Although, to be realistic, this would work like a backwards strait: that is, an army in Egypt or Sinai should be able to block naval passage. Not that there's any way to code that, unfortunately.Hive said:
Note that this is not horrifically ahistorical during some time periods. There was a Suez canal in classical times, although it went from the red sea to the Nile Delta. It fell into disrepair and was unnavigable by the 700s.
kinda off topic, but wouldn't that only have been able to service galleys? shallow galleys? nothing nearly so formidable as the ships of the mid to late eu2 time period, i'd imagine.
-Matt
-Matt
TheLoneTaco said:
It is actually uncertain if the canal was completed in 600 BC, but when Dareios reopened/completed it in 500 BC the canal was large enough to fit two large galleys side by side, galleys large enough to sail on to Persia..
Small update
Working on a tool to export province "images" to a png, which you can edit, and to import it back into the map.
I'm implementing the importing procedures, but it's a pain in the ass.
Main purpose: this will allow you hillbillies to edit the names on the provinces. Won't be possible to change province shapes.
Working on a tool to export province "images" to a png, which you can edit, and to import it back into the map.
I'm implementing the importing procedures, but it's a pain in the ass.
Main purpose: this will allow you hillbillies to edit the names on the provinces. Won't be possible to change province shapes.
Nice one. You are teh roxxorzInferis said:
I am still a bit stumped on rendering incgrid. With the TI there is just the problem that the raw data alone seems inconsistent with what's shown on the screen. With the sea provinces and some PTI areas the fog seems to roll out over the province borders, while some smaller inland provinces are almost entirely visible under the fog. I am thinking that EU2 must use some other data (like terrain type for instance) to decide how strongly to fog a province. I just can't put my finger on it.
With the lightmap encoding I see an obvious solution for lossless compression, but it will be really slow. Like unbearably slow. I haven't tried to code it yet, but it's been floating around in my head (which has a built-in predictive profiler
).
Inferis said:
Pfft, weak. I remember when you promised a full-working editor being ready for september...
Okay, fine, I'll make the program so that it won't work on your pc.Hive said:Pfft, weak. I remember when you promised a full-working editor being ready for september...