Is it always possible to get every instrument to max multiplier?

[thatmarkguy]

Take On Me is yet another unmaxable. If you hit every Vocals note before the first checkpoint you still only get to 2x, and there is no overdrive before the first checkpoint so you can't Road Rage / Bottle Rocket second-segment notes before the first checkpoint.

[Epsilon82]

"Indestructible" has no notes at all before the first checkpoint.

I played that song recently, but I can't remember if there was anything weird with the multipliers on that one. Does it give +0 or still give +3? I suppose it doesn't really matter either way as far as scoring goes, since obviously everyone would be working from the same start. The only thing it might do is make Gold Stars tougher, if for some reason the formula assumed being able to get +3 on the first checkpoint and it gives nothing, but I don't think I had any trouble getting Gold Stars on my only attempt.

[Groudon199]

I played that song recently, but I can't remember if there was anything weird with the multipliers on that one. Does it give +0 or still give +3? I suppose it doesn't really matter either way as far as scoring goes, since obviously everyone would be working from the same start. The only thing it might do is make Gold Stars tougher, if for some reason the formula assumed being able to get +3 on the first checkpoint and it gives nothing, but I don't think I had any trouble getting Gold Stars on my only attempt.

It gives +0.

[Epsilon82]

If some track multipliers can't be maxed out because there aren't enough notes, well then that's useful information to know. Does that make it impossible to get a gold star on the song? As long as it's not an in-game song (for which you "need" to get a gold star, to get a trophy), I don't see how that's a problem. It's part of the game, discovering what can and cannot be done. And, even then, people seem to have found a way around this problem: use a power-up to clear future notes (like a rocket?). So it's not a problem at all, it just required a little creativity, and it took all of a day or so.

I agree that it's not as much of a problem if it's just happening on random DLC songs rather than core Blitz soundtrack songs, but it's still a problem and should be addressed if possible. I think players have a reasonable expectation that their star score reflects the proficiency of their performance relative to the scoring potential of that song. If it's unduly difficult to get gold stars even on a virtually perfect run, and it's possible to address in the formula, then it should be done.

I'd note that this would not be unprecedented for Harmonix; they patched the original Rock Band for an issue similar to this (I believe the issue there was that the base score was attributing an unrealistic BRE bonus, making gold stars much tougher to get on certain songs because of it.)

However, after thinking about this for a while, it's not immediately obvious that there's an easy fix for it because of the nature of the game's scoring that differs tremendously from traditional Rock Band. Let's assume for a moment (and I think this is a reasonable assumption) that the game figures out what the theoretical maximum multiplier for each lane is and then uses that as a sort of base to determine what a reasonable score requirement would be for each star level. Now, my initial thought was that maybe the formula is simply counting the number of checkpoints and assuming that each of those could add +3 to the multiplier, and that maybe they weren't checking for what the true actual max multiplier would be based on the note contents within each checkpoint. This could result in the game assuming a higher max multiplier and then skewing the total score requirement.

However, it's entirely possible that they're already doing that (and that would be pretty trivial to do, honestly), but it still doesn't matter, because what's actually more important than the total maximum multipliers at the end of the song is what the maximum multipliers are at each stage of the song. For instance, it's one thing to have the game think that the max multiplier is x21 when it's really x18 or something like that. But the actual impact on the scoring is much greater if 2 of those missing multipliers happen at the first or second checkpoint, because then the lower scoring cascades all across the entire rest of the song.

So it's possible that this is way more complicated than just accounting for individual lower multipliers; you'd need to account for when they happened and then calculate the "spread" effect, which would not be nearly as simple.

Of course, I could be completely wrong about my assumption of the maximum multiplier impacting the formula this directly, but given my experience so far, I think it fits. As of now, I've played 265 unique songs (almost all of them only once with Road Rage/Blast Notes/(mostly) Super Guitar) and I've gold starred 250 of them. Of the 15 I haven't, the common trend seems to be a combination of either the song being longer in length (which I believe makes Jackpot more valuable and exacerbates early checkpoint misses) or have early checkpoint issues, or both. On none of them did I feel like my actual performance was significantly better or worse than any other song.

But there's definitely a problem if I am ranked 3rd overall on Keep Yourself Alive, 13th on Hot Blooded, and 14th on Pour Some Sugar on Me (Live) and don't have gold stars. It's just a matter of how difficult it would be to address compared to the more simplistic methods required by traditional Rock Band games.

[Meat-Popsicle]

No, you're right, it's totally "fair" to expect that gold stars should be attainable on every song, even if some wind up being (by the nature of the song) much more difficult than others. It probably shouldn't take a flawless play-through using the ideal power-ups AND an optimal path AND having the random elements to the power-ups work out perfectly to get gold stars. If there are some songs which make you work harder to find that perfect path, or require a creative use of power-ups, that would be an interesting challenge. But I suspect it's more likely that the formula for the gold star level just needs to be tweaked on some songs, and that there was no plan by Harmonix to make the players work extra-hard to solve a baffling puzzle for certain specific songs. I think you're right that Blitz is acting as if it expects the player to get checkpoint multipliers that aren't actually feasible (at least not in any regularly reproducable way). Ideally that could be regonized as a legitimate flaw and patched, as it was in your example with the original Rock Band game and unrealistic BRE score expectations.

Games used to have all sorts of "unfair" expectations of super-human performance to clear certain stages. But those were different types of games, and this is a different era, and thus I don't think it was ever intended that some songs would require 20 play throughs by a top-level player (using one specific set of power-ups) to get a gold star. Then again, if an in-pack song is one of the tracks affected ("Shout"?), that must have been tested many times, and it must have been decided intentionally that some songs would turn out to be incredibly tricky (but not impossible) to Gold Star in their Blitzified format.

[thatmarkguy]

I'd note that this would not be unprecedented for Harmonix; they patched the original Rock Band for an issue similar to this (I believe the issue there was that the base score was attributing an unrealistic BRE bonus, making gold stars much tougher to get on certain songs because of it.)

But they never patched RB3 when they had similarly unrealistic GS threshold because it assumes guitar-solo FC (the GS threshold for Green Grass and High Tides went up by 220,000 between RB2 and RB3). People always said "this will get fixed/adjusted in the next patch" ... but it never did.

They also never patched Beatles even after realizing some GSes (like "I Feel Fine" on bass) were literally unattainable.

[Meat-Popsicle]

But they never patched RB3 when they had similarly unrealistic GS threshold because it assumes guitar-solo FC (the GS threshold for Green Grass and High Tides went up by 220,000 between RB2 and RB3). People always said "this will get fixed/adjusted in the next patch" ... but it never did.

They also never patched Beatles even after realizing some GSes (like "I Feel Fine" on bass) were literally unattainable.

True, but those titles quickly reached a point where they couldn't be patched any further, for reasons that had nothing to do with "Harmonix doesn't care"*. It's been suggested (and now, shown) that this game is far easier for them to tweak - in fact, it doesn't even seem to require a patch, techincally?

* My informed guesses = the patch process for full disc games seems to involve a huge input of time and money for programmers and testers, both of which were in much shorter supply once Harmonix went independent. Also, once sales of a title slow down, and other projects gain steam, further patches on the old title become an inefficient use of resources, however much the people there might want to get everything right (perfect is the enemy of good).

[Epsilon82]

But they never patched RB3 when they had similarly unrealistic GS threshold because it assumes guitar-solo FC (the GS threshold for Green Grass and High Tides went up by 220,000 between RB2 and RB3). People always said "this will get fixed/adjusted in the next patch" ... but it never did.

They also never patched Beatles even after realizing some GSes (like "I Feel Fine" on bass) were literally unattainable.

True, but RB3 also didn't sell very well, unfortunately, and I doubt the Beatles fared that much better. At a certain point there's a level of diminishing returns for incurring the expense required to patch a game, and unfortunately that point was a lot sooner for RB3 than we'd have hoped. I can only hope that Blitz, with its cheaper price point and lower barrier to entry, may be different.

[Epsilon82]

True, but those titles quickly reached a point where they couldn't be patched any further, for reasons that had nothing to do with "Harmonix doesn't care"*. It's been suggested (and now, shown) that this game is far easier for them to tweak - in fact, it doesn't even seem to require a patch, techincally?

Well, the reason they didn't require a patch to tweak the coin economy was that everything related to it is handled server side (which is nice from that standpoint but bad from the standpoint of the user needing an uninterrupted connection to their servers to get anything actually recorded.) There are an awful lot of things that people have been discussing that would almost certainly require a software patch, for which the process and expense would likely be similar to that of patching a disc game.

I highly doubt that a fix of this nature could be handled server side, because we're talking about something extremely fundamental to the gameplay. I can't see them literally calculating star thresholds on the fly on their server every time an individual user is playing a song. I haven't actually bothered to play anything in Offline mode, but I strongly suspect that the star cutoffs are present there (not that anyone would ever be able to find out past 4 or maybe 5 stars anyway without powerups.)

And as for Shout, that is indeed an interesting edge case. Maybe they just assumed that everyone shooting for Gold Stars would be using Jackpot and so their testers didn't have too much difficulty with it. I know I got it on my first try because I heard in advance that Jackpot + guitar sustains was the key, and I didn't even play very well at all. Or perhaps whatever adjustments they attempted to make to the formula had unintended (and more severe) consequences elsewhere. Like I said before, the cumulative multiplier mechanic in this game is really interesting, but also adds an enormous amount of complexity when compared to the simpler scoring mechanics of previous games. It also invariably leads to some seemingly arbitrary issues like sustains being worth so much more than regular notes and hitting more notes at the end of the song being so much more important than at the beginning. But a lot of these things are unavoidable without scrapping the game's core design mechanic, and it's possible that what they tried to do to alleviate it ended up doing things like making shorter songs WAY too easy to gold star or something like this, and this is the best balance they could strike.

There's just no way to know, but I'd hope they could figure something out, even if it's far from the end of the world. It's much less important to me than the fact that the game handles hiccups with their server connections so poorly, for instance.

[SirDavidTLynch]

True, but RB3 also didn't sell very well, unfortunately, and I doubt the Beatles fared that much better. At a certain point there's a level of diminishing returns for incurring the expense required to patch a game, and unfortunately that point was a lot sooner for RB3 than we'd have hoped.

Rock Band 3 had something like 5 patches, the last of which was more than a year after release.