Nintendo Switch (MEGA39’s/MEGA MIX edition)

Just noticed something awful: The stage loading time on mega39s is absurdly long. Decorator takes an awful 38 seconds to load. Even simpler songs like Nostalogic can take 20 seconds. Take PS4 and my Project DIVA rig for comparison. Both store the game on SLOW 5400rpm HDD and yet they managed to load the Decorator stage in 15 ish seconds and that's VERY long already. Even the PS Vita version loads all the stages in reasonable amount of time (~10-15 seconds).

 

The famitsu stream.

 

OK then let's wait and see, internet spread too much different visions.

 

Yes, I checked on the official HP http://info.miku.sega.jp
It seems good, now let's wait to have Extreme and ExExtreme in our hands.

 

Well I'm bored so here goes another post about mega39s.

All loading screen shown on Famitsu stream timed (very unscientific stopwatch job):

Code:

Song                    M39     FTPS4   M39/FTPS4
39 Music                17s     new     -
Ohedo Julia Night       9s      new     -
Teo                     15.5s   new     -
The World is Mine       7.5s    2.75s   2.7x
Weekender Girl          17s     7.5s    2.27x
Roki                    11s     new     -
Nostalogic              21s     7s      3x
World's End Dance Hall  20s     9s      2.22x
Erase or Zero           24s     10s     2.4x
Decorator               37s     16s     2.31x
Catch the Wave          14.5s   new     -

(From selecting the start game option to in-game screen on Arcade Mode or IMU calibration screen ("The rhythm game starts now" prompt) on Mix Mode. FTPS4 measurements are taken from FT PS4 version running on a PS4 Pro with stock 1TB 5400rpm HDD. According to my experience arcade performs roughly the same and due to OS level data caching can perform even faster after repeated loading/unloading. Putting the game on an SSD might also help a lot on both PS4 and arcade)

In conclusion, mega39s takes about 2.25-3x the loading time compare to the PS4/Arcade counterpart. It can vary between minor waits to absolutely absurd depending on the complexity of stages (or anything that doesn't load within 8 seconds on PS4 takes an eternity on Switch).



(Video not related)

 

Also: from the data you can sorta see how cheaply those new PVs are made comparing to the old ones. I bet Catch the Wave PV is just a video background and an empty room with only modules and a polygon square. (Just saying)

 

Still no info about the subtle-yet-broken slider change. I guess everyone just accepted it silently.

I certainly hope so because I will likely still pick it up once it hits the bargain bin or I can find used copies (like what I did with DIVA X PS4). However performance issues can sometimes be very difficult or even imposible to fix. See for example Link's Awakening lagging issues. Some of the lags are there due to DDR4 memory being badly optimized for GPU use and unless Nintendo rolls out a Nintendo Switch SUPER (looks at Nvidia) those lags are not going away. Not to mention that if I'm going with bargain bin or used copies I will be stuck with carts and the transfer speed of carts are TERRIBLE compare to good microSD cards, which might add extra time on top of the already absurd loading time (although funny enough I tried to throttle the disk transfer rate to 25MB/s on my Project DIVA rig and the loading time stayed mostly the same so idk if the transfer speed really matters in this case).

 

What issue ? I tried it at Miku Expo so maybe I can tell you my feeling, even if I'm almost sure that it was the old TGS build in lot of aspect (but translated ? maybe not finally :/ )

For now, all I can say about sliders is that they are not mapped to bumpers from scratch in this demo, you should use joysticks by default.
The slider seems to act the same way except that little joystick make the feeling like "faster", but it's the same for a lot of games on Switch, like Sliding in Groove Coaster feel more... "attacking" the slide than a DS4 joystick ^^
But nothing weird to me

 

It is the slider spped. On PS4 you can actually push the stick briefly and let go to complete a chain slide. On switch however they decided to change the slider to fill a LOT slower compared to PS4 (36.7% the PS4 speed to be exact). While this is subtle I can't really see a reason for this to be implemented. Slides on stick is NOT pressure-sensitive to begin with and newcomers aren't gonna feel any different anyway. Also changing the speed destroy the flow of the game play at least for me because I was already used to the PS4 mechanics where I can just hit the slider and focus 100% on the next note/combo. On Switch however I need to focus on the slider way too much and it feels weird at least and kills perfect runs or makes pass/fail difference at worst.

 

BTW @cyberkevin I do have a question for you: you mentioned on Twitter that SEGA/HORI didn't fix the slider "being too fast" on the official controller. What does that mean? Did you mean that the slider is not behaving as a linear encoder like it was on the arcade? (fyi I frame-stepped the slider scene on Shake It and seems that every chain slide notes always fills up consistently in 3 frames, which was the PS4 speed (4 slots takes ~2.5 frames to fill))

If that's true SEGA and HORI would be my best scammers in 2020 as they penetrated all my BS meter and fooled me. (jk but also semi-serious)

 

For the Hori slider, I'm saying that it act the same as the DS4 touchpad: doing the slide on just a few centimeters is like doing 3 times longer distance on real arcade clearly in the last video we saw it (third songlist trailer on Twitter I guess ?)
So for now, I think they changed absolutely nothing about the way the game handle the slider of this official controller, the game don't give a f*** and just act like always, like Divaler, like all projects we see, like everything ^^

But maybe not, that's just what it look like in the trailer :/


For the Mega39's slider, I don't know...
I don't remember feeling anything different about slide completion on this MikuExpo build, but it's possible, and to be honest, it will be way better to mimic the feeling of completion and the needed time to do it on real arcade.
Wait and see :/
Thanks ^^

 

Well they admitted that they are using analog stick which is suboptimal since it doesn't represent speed properly i.e. no matter how slow or how fast you swipe it the slider always increments at the same speed. Prehaps that's what you saw?

BTW a single swipe on PS4 touchpad fills 20 slots and simply stretching the touchpad to a longer slider won't recreate the arcade behavior properly (since arcade does 31). That's why I got sophisticated setup (i.e. ATRF and TTRF as I mentioned in the controller post) to get around it.

Well depending on the situation I might just zip through the slider at super sanic speed by hand or cheat using multiple fingers. It depends on how fast player swipes it through so it's controllable. However on PS4/Switch player has no control over the speed so default fast>default slow IMO. Also why do they change that when we already got used to the fast PS4 slider speed? It just doesn't make a lot of sense.

UPDATE: Hm I properly frame-stepped the controller video (instead of only play it on slowmo and rely on my brain to fill in the rest) and it's actually suprisingly hard to tell if the controller has a native slider mode.

What I discovered:

1. The controller footage is 7 frames earlier than the game footage. Hopefully this is an editing error and not actual game lag.
2. The guy is swiping the slider at roughly PS4 slider speed, which then leads to
3. The LED pattern actually looks somewhat consistent compared to the slider increments on-screen. The video is heavily compressed and the guy's hand sometimes blocks the LED so it's difficult to draw a conclusion. However there's still hope that SEGA and HORI didn't scam us :D