FTL
Well-known member
Ok, it looks like you're really trying to understand as opposed to some others, so here goes.
Well, for a start they wouldn't be called bits as that name is specific to base 2 and it's really just a name. For base 10, they might just be called digits, I dunno. For a binary cell, it would hold just one of two voltages at a time, say 0v or 5v. For a base 10 cell, it would hold one of 10 voltages, eg 0v, 1v, 2v, 3v, 4v, 5v, 6v, 7v, 8v, 9v for example. That's the only difference, end of story.
Now, as I explained, the computer has to read those voltages and the more levels there are, the tighter the tolerances have to be and the more that they're affected by environmental factors like ambient temperature, making the job harder. Therefore, in practice, base 2 is the best base to use as it's so much more robust since there are only two levels to differentiate. I did explain this already, in a less detailed form before in this thread, but it looks like you didn't understand.
When it comes to SSDs, it's the same thing and it results in more storage per cell, but again, the more voltage levels, the harder it is for the comparator circuits to tell which voltage level is meant to be which and voltage drift really matters. Here's a big Wikipedia article on SSDs and I've quoted the relevant part to this discussion:
en.wikipedia.org
As you can see, the more levels there are, the less reliable the SSD becomes, but the bonus is much increased storage. It doubles for every level added, so you can see the attraction. Modern technology helps to mitigates the drawbacks of multi level cells, but can't eliminate them.
That's (binary [base 2]) bits per cell. You said, and I quote:
And I simply asked:
So, how would your base 10 computer store bits of data?
Well, for a start they wouldn't be called bits as that name is specific to base 2 and it's really just a name. For base 10, they might just be called digits, I dunno. For a binary cell, it would hold just one of two voltages at a time, say 0v or 5v. For a base 10 cell, it would hold one of 10 voltages, eg 0v, 1v, 2v, 3v, 4v, 5v, 6v, 7v, 8v, 9v for example. That's the only difference, end of story.
Now, as I explained, the computer has to read those voltages and the more levels there are, the tighter the tolerances have to be and the more that they're affected by environmental factors like ambient temperature, making the job harder. Therefore, in practice, base 2 is the best base to use as it's so much more robust since there are only two levels to differentiate. I did explain this already, in a less detailed form before in this thread, but it looks like you didn't understand.
When it comes to SSDs, it's the same thing and it results in more storage per cell, but again, the more voltage levels, the harder it is for the comparator circuits to tell which voltage level is meant to be which and voltage drift really matters. Here's a big Wikipedia article on SSDs and I've quoted the relevant part to this discussion:
Lower-priced drives usually use quad-level cell (QLC), triple-level cell (TLC) or multi-level cell (MLC) flash memory, which is slower and less reliable than single-level cell (SLC) flash memory.[93][94] This can be mitigated or even reversed by the internal design structure of the SSD, such as interleaving...

Solid-state drive - Wikipedia
As you can see, the more levels there are, the less reliable the SSD becomes, but the bonus is much increased storage. It doubles for every level added, so you can see the attraction. Modern technology helps to mitigates the drawbacks of multi level cells, but can't eliminate them.
You're trying to defeat what I'm saying with a strawman argument here, as I've never said that computers based on anything other than 2 were mainstream or were better in a practical sense. I simply explained that they can be built and work and that they have been created and do work.Ternary computing is still theoretical, and the last successful project was Setun; the only recent project is by someone who has not provided an update since 2016.
Everyone who has researched ternary computing (as it is still mostly theoretical, as there have been no practical systems) has stated that to use modern hardware or platforms would require a translation layer... Which would support what @MySiteGuy is saying in that current computers only work with base 2.
For someone who's supposedly such a professional IT person with decades of experience, you're having quite a lot of trouble understanding basic concepts.If you're going to argue theoretical semantics, do so with someone who is not a professional IT person who has decades of experience and more familiarity with the subject that you're arguing about.