I've been playing around with the Opus codec today. It's working, but it's funny how those comparisons don't mention that decoding is 3-5 times slower than vorbis.


3-5x slower doesn't sound right to me... are you using special implementations? With off-the-shelf tools I find opus is usually a ...

One way you're making things hard for yourself is that you're fighting physics: there is essentially no chance of this working if you use a lossy codec like vorbis, due to the artefacts you're introducing in the frequency domain. Getting a real codec / player to support gapless playback with one ...

Minor graphics issue.

In one of the FFF-387 videos I noticed a vertical cut in both the copper plate and plastic/steel belts and it turns out this is an inserter hand shadow. There seems to be about a 30-degree angle between the shadow and the line of the jaws; this isn't necessarily wrong but I ...

It looks like after you placed the long belt section, you kept the LMB pressed and tried to use smart dragging to rotate the belt on the RHS in order to join it to the 2-tile belt. However I think smart dragging is only meant to work from the last-placed belt, which is the one on the LHS in this ...

I think it is - actually your post motivated me to chase down this problem of the range being inconsistent. I'll post a comment.

Minor issue.

It is possible through normal play to set the smart belt 'snap dragging' range to either 10 or 20, and this will persist through building attempts.

Firstly the setup to test the range:

Place a vertical belt and keep the LMB held down.
Walk roughly 20 squares right of the belt ...

I tried minifying the Ximoltus save file and ended up with this example that's pretty far removed from the original. Here are three copies of the same blueprint; the speaker is set to beep when the belt has < 4 iron.
three.png


A looks pretty normal and the speaker is silent.
B shows some of ...

It is possible for cliffs to reach what looks like an inconsistent state, in which they are visually marked for deconstruction with a red cross but do not appear to have any current or planned deconstruction orders: Bots ignore them, and the red cross can't be removed by shift-selecting a ...

This happens to me sometimes (1.1.104).

Best guess at the condition to trigger: The grid position coords must be odd, and the horizontal distance from the leftmost tile/entity to the leftmost rail entity must be odd. Similarly for topmost.

So in your posted blueprint, I think it's the topmost ...

Note I believe this report is technically a dupe of 93236 . To repro that issue, here is a combinator with a checkmark on it

blurry-check.png

I believe the reason the glitch is happening is an incorrect mipmap (possibly the only bad one in __base__/graphics/icons). The two levels are ...

This is a minor but noticeable rendering issue that arises from the combination of two non-bugs:
Icons have flags 'linear-minification', 'linear-magnification', 'linear-mip-level'. So if the game interface has a UI scale that results in icon sizes between mipmap levels (i.e. UI scales 75, 125, 150 ...

I use a lot of unreachable train stops on purpose - it's the only way I know of to say "wait until next stop is available". From reading other posts, this appears to results in train repathing performance issues.

With this setup, if I mark one of my train grid squares for destruction then the ...

I suggest you use train limits instead of unreachable stops. I don't see a large drop in performance with that map, but I do have a fairly beefy computer, and there is a small drop in perf when marking one of the intersections for deconstruction. The drop goes away when I replace the unreachable ...

I use a lot of unreachable train stops on purpose - it's the only way I know of to say "wait until next stop is available". From reading other posts, this appears to results in train repathing performance issues.

With this setup, if I mark one of my train grid squares for destruction then the game ...

I found some odd behaviour when I had a quick glance at the map while drag-placing a long stretch of belt. Minimal steps to reproduce:

Left-click to place a single belt and hold the mouse button down
Press M to open the map (keep the mouse button held)
Move the mouse a few tiles in the direction ...

bpflip.png

Create any blueprint, e.g. A.
Horizontally flip the blueprint while in hand, no need to place it anywhere.
Update the blueprint's contents with an unflippable blueprint, e.g. B.

The result is C, which is invalid (the station is the wrong way round) and cannot be flipped back to B ...

FWIW I added a short 'on_entity_settings_pasted' callback to my copy of IR2 to sanitise paste results. So you can copy nearness / farness between short / long inserters without changing their lengths. It doesn't feel quite right but it seems to work.

Yay, specifying an almost-arbitrary centre of rotation is now possible in 1.1 by using an absolute offset. My train set rotates and tiles properly now. Thanks devs

So how about another way - use current behaviour, but add an "offset anchor". Show the "center map" in the blueprint GUI (intersecting lines like in the picture above, with some flag in the center) and allow to drag it (e.g. by Shift + RMB). Unchanged anchor will automatically align to the corner ...

Blueprints will not change their "offset class" when rotated.


That's true in the cases we've both drawn, but when a rotation acts on the set of offset classes, (0,1) gets swapped with (1,0).


... it can be easily fixed by adding some controllable offset to the blueprint tile grid.


Are you ...