PORTLAND TIMING IN GTA III EXPLAINED _

Doc by oasiz, contact info at the bottom. Some additions by Lighnat0r.

You need to do portland as fast as possible but there is one catch! The first mission you do in Staunton (2nd island) where you kill Salvatore is timer based.

Let's break it down:

• Salvatore's exit happens only every xx:30¤ based on the ingame timer.
• A good first target would be 15:30 portland as you can still do a couple of mistakes and not have the best RNG and still get it. Once you think that you are tough enough then go for a 14:30. 13:30 is possible but requires an insane amount of dedication and practice to achieve, not to mention luck.
• When you trigger the mission the game does the following: ¤ Rounds up to the next minute by adding each remaining second, let's say that you trigger at 12:40 -> 20 seconds ¤ Adds 2 minutes and 30 seconds to the above value, which is 20 seconds in this case. -=> Salvatore leaves at 15:30 and you have 2:50 to waste.

Bottom line: The more time you have left to the closest ingame minute (faster) the more you will wait for salvatore.

How does this translate to portland performance ?

You start the game at 4:00~ ingame time roughly where the first 0:10 is spent on Eight ball talking about stuff

According to the math, you need to finish portland before 13:00 on the next day to get a 15:30 salvatore ! This translates to roughly 32:50 of gameplay in portland (ingame clock).

Yes, this means that no matter how perfect your portland is, you absolutely must fit 'salvatore exit' in the time slots (see below).

== THE MASTER TABLE OF TIMES (with cool advice!) ==

TIME | DIFFICULTY | WHAT DO THE STREETS TELL YOU | NEEDS _ 19++ - CASUAL | Just keep playing a bit more! Soon you will be getting 18:30 times | More Soy sauce 18:30 - BEGINNER | Starts to be optimized, requires some skill already | Basic route, OK RNG and gameplay 17:30 - EASY | Mostly good runs with no major mistakes, goody ! | Doable even with OK RNG and skill 16:30 - NORMAL | Getting pretty good here with this, already enough for 1:25 final times | Decent skill needed if bad RNG 15:30 - CHALLENGING | Prepare to restart a lot, you are already a veteran Portlander with a solid route | Really good skill and OK RNG 14:30 - REALLY HARD | This should only be attempted if your current best time is below 1:21:30'ish | Lots of skill but even more RNG 13:30 - INSANE | Nothing left to master in portland. If you manage to finish that run, bravo! | Godlike RNG and execution. RARE. 12:30 - IMPOSSIBLE | Nice TAS you just did. | Execution like no one can do.

A typical 14:30 portland can have up to almost a minute to waste with a really good run so small mistakes can add up really easily and aren't tolerated because of that. Even with a otherwise well executed "on track for 13:30" run, RNG plays a big role.

This all is a blessing and a curse at the same.

• You can do 5-10 seconds worth of mistakes and they don't mean a thing.
• RNG isn't super dependant all the time.

• You MUST improve by up to a minute to get to lower minute for salvatore to come out faster.

There have been numerous changes on the "good time" over the years. At one point 15:30 was pretty hard to achieve and 14:30 was barely realistic. Few critical strategy changes shifted the timer down by almost a minute, making 14:30 possible and 13:30 about the same what 14:30 used to be. While 12:30 is possible in theory, it will require so much "once in a lifetime" luck to get it. If you combined the best mission from every run done by a good runner (who gets 13:30), you would probably end up with 12:30 portland time.

There is, in fact, a way to manipulate the ingame timer, allowing you to get an earlier salvatore exit at the expense of some time. Internally, the game keeps track of "ingame seconds", so that 60 of those add up to one ingame minute or one second IRL. Every time you start a replay (PC only), the ingame seconds are set to 0. This means that by timing correctly, you can keep the ingame timer from progressing altogether by starting and ending a replay every second. As you can imagine, doing this constantly just wastes a bunch of time. But if you would normally miss 13:30 by 5 seconds (which would have you waste 55 seconds since you'd get a 14:30), you can use this glitch a couple of times to get the 13:30 (costing a couple of seconds, but saving those 55 so overall it's faster). While a 13:30 with this glitch is evidently slower than a 13:30 without it, it is faster than getting a 14:30 without the glitch would've been.

¤ One ingame minute in the clock equals one second IRL. So time passing from 1:30 to 2:44 would mean 1 minute and 14 seconds.

• • • • rev. 3

    "I don't understand anything you said, and you are a shithead!" Did you know that you can say it straight to my face? Catch 'oasiz' on IRCnet/Quakenet/Freenode. I'll gladly help you if you have any questions about this route. And I stream TekWar from time to time at http://twitch.tv/oasiz On Speed demos archive I am known as 'oasiz', also known as the one who did the GTA3 run at AGDQ2013