It completely depends on how much you are willing to practice.
Whichever system you are willing to train to fluency.
Not to be sarcastic or dismissive, but there is no way to answer this question for someone else. Every individual is different and a technique that works great for one person may not click at all for another. Only you will be able to determine which system is “best” for you. The only way to do that is to experiment and actually try to use them for a while.
Don’t worry about picking the “wrong” system. You can always switch later. The brain is pretty adaptable and previously learned associations can be changed or overwritten with a little bit of work. When I started out, I learned a category system for playing cards, unrelated to numbers. Then I switched to basing both my cards and numbers on a Major-based PAO. Then I changed the way that I associated specific numbers with specific cards. Then I learned a 2-card two-block system and 3-digit number system. Then I changed to a PO 2-card system. Finally (I hope), I changed to a true 2-card system and a 3-digit number system with modified strict Major associations. Over the course of that evolution of systems, I overwrote probably over 2000 associations and re-trained them all to be fairly fluent. I’d be willing to bet that if you stick with this, you’ll eventually want to adjust things or change or adapt systems. So don’t stress it now, just get started.
I’m not totally sure what you mean by this. Are you talking about a 3-digit system with a single association per number vs. a 2-digit PAO? This has been endlessly debated on the forum. Here’s my take on it…
(WARNING: LONG RESPONSE INCOMING! haha)
For me, the “upside” of a system always has to do with efficiency of data compression and how many digits or cards or whatever you can represent by a single intentional element. This phrase is important. I think of “intentional elements” as those that actually encode information and that you must actively consider and translate accurately when memorizing and recalling.
In a 2-digit PAO there are three intentional elements that you must recognize, translate, and integrate into a scene. The Person, the Action, and the Object. Each of those must be “read” from the numbers. Each of those elements compresses two digits down into a single intentional element, for a total of 6 digits per scene.
For a 3-digit single association system, you can still get 6 digits in a scene, but the crucial difference is that those 6 digits are encoded into just two intentional elements. You only have to actively consider or “read” two things in order to construct your scenes.
With both systems there will also be incidental elements in your scenes. These are details that don’t encode any information but serve to make the scene coherent or add memorable aspects to it. These might be things like facial expressions for your people or the effect that the actions in the scene may have on the location. These details don’t really “count” when it comes to evaluating the complexity or efficiency of a scene. They’re just there naturally without any active effort to construct them or purposely place them within a scene. If your intentional object is a “car” to represent the number 74, then some incidental elements might be the mirrors, windshield wipers, steering wheel, door handles… Even though those are technically their own additional objects, and you will probably visualize them as part of your car, they don’t actually encode any additional information, and they don’t really require any thought to include in your scene, they are just “there” as part of the car.
To bring it back to the PAO vs 3-digit question:
Lets say I want to memorize 123456. I can use my PAO and memorize it as 12-34-56 (Tony Marrying a Leash.) If I do that then I need to consider each of those three intentional elements and make sure that I accurately and purposefully construct the scene using them. In this scene, I may imagine a priest standing there performing the ceremony and Tony may be dressed in a tuxedo and the leash in a wedding dress… Those embellishments are incidental details that just naturally flow from having to imagine “marrying” as an intentional element of the scene.
If I use my 3-digit system, I memorize it as 123-456 (Tin Man - Relish), this is only two intentional elements to have to consider. I can imagine the Tin Man from wizard of oz pouring a huge bucket of green relish over his head. In this case the bucket is an incidental element and the action of him pouring it on himself is also incidental, it was an effortless outgrowth of combining those two intentional elements to form the scene.
I can store the same amount of digits with the same amount of scenes across the same amount of locations with both systems… BUT, with my 3-digit system, I have 33% less intentional elements to have to consider. There is where the huge advantage is. I can go faster and I have less total elements to recall in order to accurately reproduce my series of numbers. To memorize 60 digits, I only need to consider 20 intentional elements. A 2-digit system will require 30. Thats a significant difference.
The trade-off, is of course, that I need to learn 1000 “intentional elements” with a 3-digit system, compared to 300 with a 2-digit PAO. It will take more time and effort to get the 3-digit system to fluency and it will be more work to maintain that with regular drilling. Whether that trade is worth it is up to you to decide.
This depends on how you’ve structured your card PAO, but I assume that your card system is currently not compatible with Major? The cool thing is that you can easily map cards onto numbers (and vice versa), especially if you’re using a number system based on Major. There are many threads on the forum detailing all kinds of ways this can be done.
Unless you’ve worked with your card PAO to the point where you’re consistently able to memorize a deck in under a minute, it would probably be worth looking at reworking your card system to map to whatever number system you end up going with. This way you have a single list that applies to both and as you get faster with recognition it will help both.