What's the most efficient way to produce human IPSC?

Every paper and every product wants to claim it's the best, so I'm interested to hear what systems other people are using to generate human IPSC.

In my experience, I'm always trying to balance cost, efficacy, and safety, but nothing seems to have the optimal balance. I'm currently using Plat-A-generated retroviruses encoding Oct/Klf/Sox/Myc plus valproic acid, which is really cheap and quite safe, but the efficiency of IPS generation is awful mostly because the retroviral particles are weak and unstable.

I'd like to switch to more efficient virus, such as a polycistronic lentiviral system or at least VSV-G pseudotyped retrovirus for better transduction efficiency. However, I'm concerned about safety issues with generating and concentrating stable amphotropic virus encoding human oncogenes, yet buying pre-packaged lentiviral particles is prohibitively expensive for what I'd like to do.

I've heard about people trying to create them without vectors. Have you heard much about this?

Hi, you guys! Some of you may remember me as that one guy named Sozeru who made it to Asurahood and like Level 22 or something! The lovely Auuralei directed me to this forum when I asked if there's a place for the guild to hang out together, so I finally thought "Why not?" And here we are now!