By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Gaming Discussion - Cartridges or optical discs?

 

I prefer...

Cartridges 381 78.56%
 
Optical discs 104 21.44%
 
Total:485

Neither, I've been mostly digital for a while. I don't really mind either but I always used to prefer discs just because the save data isn't tied to the game and is easy to back up. I know this could be done with carts but it never really was until the Switch and now its pointless for me since I'm digital now.

I suppose discs also used to offer superior storage but thats also a moot point now.



Around the Network
fatslob-:O said:
Pemalite said:

Switch should have an advantage on random access times though.

In general... The Switch isn't breaking 90MB/s of bandwidth regardless if it is cart or MicroSD... And because of such, the mechanical drives in the consoles are actually faster at sustained reads and writes. (The Optical Disks are much slower than all of them however.)

As do all flash based technology since they have the lower access latencies ... 

@Bold You can say that again since Nintendo hardly gives a damn about how fast their cartridges are if their willing to use monstrosity such as this ... (1-bit bus width, 45 MHz clock all for a whopping 5.6 MB/s!) 

My internet is faster at downloading than their 3DS cartridge read speads LOL ... (I wouldn't rule out Switch carts being possibly slower)

Well. The Switch's  cart transfer rate seems to be pretty competitive to MicroSD as per Digital Foundry's testing.
http://www.eurogamer.net/articles/digitalfoundry-2017-best-microsd-cards-for-switch-loading-time-comparison

So we can summise that it is something like 80-90MB/s. Nothing ground breaking, but not entirely useless either.

Spindel said:

I never said that I want a cartrige that physically looks like the 8-bit, 16-bit and N64 ones.

Fair point.
With that, even mobile carts like Gameboy/DS/3DS have a fair amount of empty space as well. Certainly enough space for a Co-Processor that does not use a heatsink and fan.

Spindel said:

What I want is the ability for co-processors (whitch will require a bit roomier cartriges than the DS, 3DS and switch ones.


The main issue is building a Co-Processor that is powerful enough over the host machine whilst not requiring a heatsink and fan.

Take the SuperFX chip in the SNES, that was orders-of-magnitude faster than the SNES CPU, but it also didn't use a heatsink or fan... We simply can't do that anymore as chips have gotten orders-of-magnitude faster and more power hungry.

Now we could *potentially* still have Co-Processors that are designed for very specific tasks such as decompression of textures, but that only becomes worth it if you are extremely constrained by the hardware which just doesn't happen as much these days.

Spindel said:


It doesn't need to be a fully blown GPU (like the Super FX and FX 2 chips). But I could see a posibility for maybe a ARM chip (which doesn't neccesairly needs a big heatsink, if it needs any at all) fore some extra processing power for thing like maybe AI (yeah yeah I know ARM are more general purpose and not suited for AI but you get the point). Or a custom chip that helps to allow for some newer generation shader that is not supported for the aging hardware in the console for crisper graphics (even if I see new chipsets for stuff that is mostly related to under the hood calculations over graphical expandability since the graphics race is tiring and in most of the powerhouse consoles and PC I feel like the big budget games spend all the budget on visuals and forget gameplay).

Well. Such a chip would be extremely slow, orders-of-magnitude slower than the host machines processor... So the performance gains would be fairly marginal.

I mean... I would *love* for that to be a focus again, it's just not going to be all that realistic, but I guess this thread is entirely about your personal preferences, not what is realistically achievable with todays technology.




--::{PC Gaming Master Race}::--

Pemalite said:
fatslob-:O said:

As do all flash based technology since they have the lower access latencies ... 

@Bold You can say that again since Nintendo hardly gives a damn about how fast their cartridges are if their willing to use monstrosity such as this ... (1-bit bus width, 45 MHz clock all for a whopping 5.6 MB/s!) 

My internet is faster at downloading than their 3DS cartridge read speads LOL ... (I wouldn't rule out Switch carts being possibly slower)

Well. The Switch's  cart transfer rate seems to be pretty competitive to MicroSD as per Digital Foundry's testing.
http://www.eurogamer.net/articles/digitalfoundry-2017-best-microsd-cards-for-switch-loading-time-comparison

So we can summise that it is something like 80-90MB/s. Nothing ground breaking, but not entirely useless either.

How do you get to that conclusion. The article clearly states it maxes at UHS-1 speeds. That UHS-1 card they used has upto 80 MB/s read speed. And the game from cartridge loads slower than from an UHS-1 card. It's not useless but below 80 MB/s for sure. Too bad they didn't test slower cards to see how they hold up.

Anyway it's not the goroundbreaking difference with discs that people seem to assume when switching to cartridges. Lego city undercover still takes 51 seconds to load on Switch, compared to 1:04 on WiiU, yet much slower than on ps4 wich loads in about 30 seconds.



SvennoJ said:

How do you get to that conclusion. The article clearly states it maxes at UHS-1 speeds. That UHS-1 card they used has upto 80 MB/s read speed. And the game from cartridge loads slower than from an UHS-1 card. It's not useless but below 80 MB/s for sure. Too bad they didn't test slower cards to see how they hold up.

Anyway it's not the goroundbreaking difference with discs that people seem to assume when switching to cartridges. Lego city undercover still takes 51 seconds to load on Switch, compared to 1:04 on WiiU, yet much slower than on ps4 wich loads in about 30 seconds.

Just throwing out rough guestimates as we don't actually have any real-world transfer rate data on-hand, something that Digital Foundry hints at.
One thing to keep in mind is that the load times between Cart and the MicroSD is extremely minimal, within a possible margin of error.

And with the 16GB UHS-1 card having a max of 80MB/s... And the fact that Nintendo also has varying degrees of ROM capacities and potential speeds... Well you get the idea. 80-90MB/s is probably on the upper-end of cart transfer rates.
Carts with smaller ROM chips will probably get significantly less than that.



--::{PC Gaming Master Race}::--

Pemalite said:

Just throwing out rough guestimates as we don't actually have any real-world transfer rate data on-hand, something that Digital Foundry hints at.
One thing to keep in mind is that the load times between Cart and the MicroSD is extremely minimal, within a possible margin of error.

And with the 16GB UHS-1 card having a max of 80MB/s... And the fact that Nintendo also has varying degrees of ROM capacities and potential speeds... Well you get the idea. 80-90MB/s is probably on the upper-end of cart transfer rates.
Carts with smaller ROM chips will probably get significantly less than that.

What SvennoJ said ... 

We really won't know what the Switch's true transfer speeds are until we start reverse engineering the system and get it to do custom benchmarks or figure out the specs of the carts themselves ... 



Around the Network
Pemalite said:
SvennoJ said:

How do you get to that conclusion. The article clearly states it maxes at UHS-1 speeds. That UHS-1 card they used has upto 80 MB/s read speed. And the game from cartridge loads slower than from an UHS-1 card. It's not useless but below 80 MB/s for sure. Too bad they didn't test slower cards to see how they hold up.

Anyway it's not the goroundbreaking difference with discs that people seem to assume when switching to cartridges. Lego city undercover still takes 51 seconds to load on Switch, compared to 1:04 on WiiU, yet much slower than on ps4 wich loads in about 30 seconds.

Just throwing out rough guestimates as we don't actually have any real-world transfer rate data on-hand, something that Digital Foundry hints at.
One thing to keep in mind is that the load times between Cart and the MicroSD is extremely minimal, within a possible margin of error.

And with the 16GB UHS-1 card having a max of 80MB/s... And the fact that Nintendo also has varying degrees of ROM capacities and potential speeds... Well you get the idea. 80-90MB/s is probably on the upper-end of cart transfer rates.
Carts with smaller ROM chips will probably get significantly less than that.

The fact that all the load times, internal, cart, different MIcroSDs are so minimal suggests that the game itself is the limiting factor. The 90 MB/s cart is even slower in some cases than the 80 MB/s cart. For all we know access speed could be limited to 30 MB/s or less and what we're seeing is the effect of different access times. Same with Lego city undecover, 51 seconds on Switch vs ~30 seconds on ps4 from HDD. Is it the read speed, or does the game do a lot of preparations while loading which the ps4 can perform much faster.

It's weird though that the WiiU version of botw loads faster after death and going in and out of shrines. Perhaps the internal storage of WiiU (3GB install for BotW) is faster than that of the Switch. http://www.nintendolife.com/news/2017/03/heres_why_speedrunners_prefer_the_wii_u_version_of_zelda_breath_of_the_wild

Anyway my point was that I'm not seeing any big advantages in load times between cartridges and optical discs (with install) Instantly ready to play is kind of a moot point too nowadays as the disc usually has installed enough to play before the day 1 patch has been downloaded.



Optical disks. Cheaper, larger capacity, more convenient - easier to store my games collection (playlist).

Carts come with extra cost and prices are affected by current NAND demand.

I see people mentioning load times. Its 2017 people.. the differences are negligible from NAND and 2.5" HDD speeds. Most games need an install on a hard drive to run on anyway.



Cartridges. Perhaps as SD Cards become cheaper, cartridges will become the standard again.



Lifetime Sales Predictions 

Switch: 151 million (was 73, then 96, then 113 million, then 125 million, then 144 million)

PS5: 115 million (was 105 million) Xbox Series S/X: 57 million (was 60 million, then 67 million)

PS4: 120 mil (was 100 then 130 million, then 122 million) Xbox One: 51 mil (was 50 then 55 mil)

3DS: 75.5 mil (was 73, then 77 million)

"Let go your earthly tether, enter the void, empty and become wind." - Guru Laghima

SvennoJ said:
Pemalite said:

Just throwing out rough guestimates as we don't actually have any real-world transfer rate data on-hand, something that Digital Foundry hints at.
One thing to keep in mind is that the load times between Cart and the MicroSD is extremely minimal, within a possible margin of error.

And with the 16GB UHS-1 card having a max of 80MB/s... And the fact that Nintendo also has varying degrees of ROM capacities and potential speeds... Well you get the idea. 80-90MB/s is probably on the upper-end of cart transfer rates.
Carts with smaller ROM chips will probably get significantly less than that.

The fact that all the load times, internal, cart, different MIcroSDs are so minimal suggests that the game itself is the limiting factor. The 90 MB/s cart is even slower in some cases than the 80 MB/s cart. For all we know access speed could be limited to 30 MB/s or less and what we're seeing is the effect of different access times. Same with Lego city undecover, 51 seconds on Switch vs ~30 seconds on ps4 from HDD. Is it the read speed, or does the game do a lot of preparations while loading which the ps4 can perform much faster.

It's weird though that the WiiU version of botw loads faster after death and going in and out of shrines. Perhaps the internal storage of WiiU (3GB install for BotW) is faster than that of the Switch. http://www.nintendolife.com/news/2017/03/heres_why_speedrunners_prefer_the_wii_u_version_of_zelda_breath_of_the_wild

Anyway my point was that I'm not seeing any big advantages in load times between cartridges and optical discs (with install) Instantly ready to play is kind of a moot point too nowadays as the disc usually has installed enough to play before the day 1 patch has been downloaded.

Hence my statement... "we don't actually have any real-world transfer rate data on-hand" :P

Installs are here to stay it seems.
As long as games don't take 20 hours to install like Halo: The Master Chief Collection it will be alright.



--::{PC Gaming Master Race}::--

Pemalite said:
Mr_No said:

Then why respond to my comment on first place if it's an opinion you care little about? Also...



Your image is non functional.

I don't care about you or your opinion, I care about facts and points that are presented and will argue against those based on their merits.

Whoa, whoa, whoa, chill. I never asked for you to care about me, but I feel flattered you let me know you don't. I argue against points too and I'm not afraid to be wrong. To be honest, I don't care about the facts and points you're meticulously arguing and raging against, I care about you obnoxiously replying me without me wanting for it. In fact, I'm gonna say it again:

I prefer optical discs over cartrdiges because they offer more space and better sound quality

And just in case, when I first came up with that comment, I had more in mind the Fifth Generation of Videogame Consoles and not the current one, even if I added terms like "tiny cartridge" and "Blu Ray Discs". How can that get into you? Back then cartrdiges were inferior in many ways to optical discs (but not in all ways). Sure, cartridges had faster loading times and had capabilities for better 3D graphics, but discs were cheaper to produce and offered excellent video and audio quality. And those are facts.

Now, you might want not to indulge in this "less than intelligent" conversation like you stated before. So don't.