I’m up to about 200,000 in a single document that I don’t want to break up, and its already going a bit slower. How many words do you think one can hold before crashing?
It depends on your computer. There is no absolute limit. It’s been tested with projects (not single documents) well north of a million words.
But why don’t you want to break up the document? Most of Scrivener’s advantages come from the ability to break a project into smaller, more manageable chunks.
Katherine
If you have it all in one single document you risk losing everything if something happens when Scrivener is doing its auto-save. If you break it up in smaller pieces, only the small document you have in the Editor risk being corrupted.
Auto-save must take quite some time if you have 200 k words in one single document…
What is your reason for not splitting it? Is it one single continuous text without line breaks or paragraph breaks?
And in case you mean document in a loose sense to describe the entire Scrivener project, and its 200,000 words scattered into a bunch of different items in the binder already—then in that case you shouldn’t be running into much by way of performance issues, there is probably something else going on. I have about 300,000 in my current WIP, and the only thing that’s really “slow” about it is how long it takes to back it up or reindex it if something goes wrong.
But if you are typing 200,000 words into a single binder item, then yeah I reiterate all of the above: this isn’t that kind of program, and it’s not optimised to work that way either. If freely scrolling around is the main thing you’re worried about losing, don’t. You can select 2,500 separate “documents” in the binder and temporarily tape them all together into a single long spool of text just like you have right now, with the View ▸ Scrivenings menu command or ⌘1.
But if you’re like me, once you get the skeleton of a monster like a 200k document described into a highly topical outline in the draft folder, chances are you’re rarely going to look at the entire thing all at once. If I need to find a section I can either drill down to it via the detailed outline—or I suppose I can load hundreds of pages of text and scroll around for five minutes. 
It’s all background material. I didn’t want to split it up at the risk of getting bugged out by all the separate documents in the folder. I color coded most of the text so I could scroll based on that, but if it’s a risk for the auto save I’ll definately split it and learn to live with it, especially since it’s still freaking growing.
Damn, I overlooked that! Am I able to move the documents around when they’re taped together?
My lack in organizational skills has always been my greatest shame 
Once you’ve broken it up into separate documents, then you can do cool things like assign keywords to them. For instance, for research you could assign keywords to docs based on Categories, or Type, or Location, or whatever sorts of info that you’re tracking.
Have you looked at the Interactive Tutorial yet? If not, I definitely recommend doing so. It’s available from the Help menu, and is a good overview of how the various parts of the interface work together.
Katherine
Yup! I’d say this whole concept right here is really the crux of using Scrivener. From the point where you’re starting, find the outline that best suits your material, chop the document up into that outline, and keep chopping if you find you weren’t specific enough to start with.
How specific? I have sections that terminate in one-liners and paragraphs. You needn’t take that far of course, my point is: the software can do that and even more—so it can handle a few 10k chunks no problem, if that’s what you’d find best.
Sure! If you move things around, split a file in two, add a new one or take any other action that impacts the contents and order of the session, it will automatically update. You can even do a lot of that right in the editor. If you press ⌘N while typing, a new section in the outline will be created after the one you were in, and you keep writing. When you move things, you will see the text shift in the Scrivenings session just as if you’ve cut and pasted it (or deleted it, if you file a section off to your “Unused” folder, thus removing it from the scope of the section while merely moving it from one place to another inside the project).
I’d suggesting going through the manual on this feature if you’d like to see what all it is capable of. There is quite a bit to it beneath the surface, like section titling, colour coding by label, etc.
Think of it this way: if your outline represents those key scroll points, then when you want to go to “blue section” you just point at it and click, if it is its own file. If the item is within the Scrivenings session you are already using (say the whole chapter), then it merely scrolls you to that point rather than loading it all by itself. On top of that, up in the top right corner of the editor toolbar, when using a Scrivenings session, you’ll see a “menu” button—by the up/down arrows and split button—click that for a session ToC.
Once you do get things split up, tools like the Outliner start to make more sense, I’d think. Adding a column to check the word count goes from giving you a total estimate to a tool that can help you find lean or fat sections, maybe shuffle stuff around to avoid having two lengthy chapters back to back, that sort of thing. It gives you a better picture of your text, even if you don’t start in with labels, keywords and all of that stuff. A simple outline flagged by modification date is fully automatic and thus a useful tool for retreading recently worked upon areas, giving them another pass.
Would a couple of millions words(series of books plus all research) in a single project lead to issues?
Not while writing, but if you want to make a backup copy sometimes while writing (a backup, not just saving) it would take quite some time to make a complete copy of the whole project and then zip it.
And without backups, what would happen if your project got corrupted? A few million words down the drain?
What would be the point of having it all in one project?
Not having to organize difference projects(plus organizing the backups/forgetting to do one…) or having to open multiple at the same time to review information. Plus, I can cross reference and bookmark…
So, the only problem is it would take a bit longer to back up?
If your research has or will have lots of big images or videos, it could take a lot longer to backup.
But if it’s just text, you should be fine.
If you do choose to keep the series in one big project, that’s an “all your eggs in one basket” scenario, so be sure you are utilizing Scrivener’s built-in backup system to make regular zipped backups.