The Varian Saturn instruments were inexpensive when new, and are inexpensive now, for a reason.
They CAN give good results, but among other things Finnigan treated the construction of ion traps as something of a trade secret, and the Varian design is a bit of a reverse engineering attempt at them. They work well in good shape with a competent operator, but need you to be a bit more hands-on than a novice GC-MS user would probably want to be.
Ion traps have pretty much fallen out of favor in GC-MS applications(they are still common for LC-MS). I talked to Thermo a year or so back about a Q-Exactive GC-MS, and they said it hadn't been made in a couple of years.
Ion Traps have some upsides and downside. Unlike a quadrupole where the quads just scan continuously, there are several "stages" in acquiring an ion trap spectrum. Even so, surprisingly enough, they can be more sensitive in full scan mode than a quadrupole, but the quad rules for selective ion monitoring(something that to me is a bit counter-intuitive).
One big issue is that traps give non-classical mass spectra. If you know what you're looking for, this isn't an issue, but it can be if you're matching up to a library.
The Varians are super easy to maintain-I can breakdown, clean, and reassemble the complete trap a lot faster than I can an Agilent source. Still, though, as mentioned, you're on your own for service.
On the other hand, in operation they are somewhat more hands-on. When tuning, with an Agilent you just open the cal gas valve and it's all correct. On the Varian traps, there's a needle valve on the front and you have to adjust it slowly while watching the screen to make sure you're getting the right amount of cal gas. Most mass specs have RF coils that can be adjusted(called "dipping" the coils). On Agilents, I've dusted off 30 year old ones, dipped the coils as part of the set up, and found that often times they didn't need to be touch, or sometimes needed a tiny tweak. On the Varian traps, I dipped the coils EVERY SINGLE TIME I tuned them, and usually they needed some adjustment.
Something worth mentioning-the Varian traps specifically, and as far as I know all traps-MUST have helium present in the trap to work correctly. Forget using hydrogen or some other less expensive carrier gas in them-they just won't work.
I find the 3800 a bit of a cantankerous GC. The EPCs are a bit buggy, especially compared even to a generation older Agilent instruments. 3800s are very capable instruments, but need to be a bit more hands on(and hope to goodness you don't want to run a small split ratio). I fought wars with the CP8400 autosampler and finally quit using it. Even my old 7673 HP/Agilent samplers can run for years without an adjustment and only an occasional syringe change. The 8400 ate syringes for breakfast, and calibration was a constant battle.
I sporadically kept our 2100 running at my last job, but at the end of the day it just wasn't worth it.
The 1200 and 300 triple quads are superb instruments, and despite the lack of service availability, I'd still recommend them. The ion traps are a different story.