Transmission made simple

[donpost]

Someone already said the theoretical optimum gear set for a specific track is the one has the highest average HP (with respect to time) over the course of a lap. This is why the higher gears are given closer ratios than the lower gears. So I guess the rules to tune by are 1) Higher gears should be spaced closer together, and 2) The gears you use the most on the track should be spaced closer together.

I made a spreadsheet from a tool I made for Forza that gives you gear ratios that make 3rd, 4th, and 5th closer together while keeping the other gears rational. You can also tweak just how much the gearing is biased towards those gears by changing a single parameter. It only works for 6 speed transmissions right now.

https://docs.google.com/spreadsheets/d/1XixkgFozPX5mhZG_DtSPNOzczLmGvQi-_a0Th1mYrjI/edit?usp=sharing


I just spent a few hours seeing if you can improve your 0 - top speed time by tweaking the gears and there I could find no improvement over default. If I went crazy I could make it a lot worse though!


EDIT: I agree about those power curves - they are hopeless! I think there is more laptime to be gained from figuring out how to find the correct shift points than there is by messing with the ratios but I am absolutely stumped

 

[SiNS3V3N]

Keep in mind tuning to make top acceleration is all good when drag racing. Tracks have corners, gears too close causing multiple shifts in corners is not good. Gears only focused on being close overlook the acceleration curve of the engine.

It's also important to keep in mind.
Close is good yes, but how close depends on the engine. A 4banger Honda that doesn't peak tq until 8300rpm and peaks hp at 8800rpm (Integra Type-R) will like super close gears to keep the power inside a narrow power band, while a V8 making peak tq at 3500 with peak hp near 6500rpm will like gears spaced much further apart compared to the Honda 4banger.

Honestly the Max Speed Slider equally spaces gears based on engine limits to reach a top speed.

That's it, done.

Now if my 3-4 shift occurs on a given track shortly before a corner, forcing me to grab 4th for a second before downshifting for the corner, we lengthen 3rd to delay the shift so I redline 3rd at my braking point. No unnecessary shifting. That's what tailoring a gear set to a track is about.

Simply moving the FD to adjust the gears to a track is no good. Sure it will move top speed, but you either make first and second too short or too tall, the same with the rest of the gears but the damage to first and second is most critical.

 

[killerjimbag]

It's PD's programming. Don't take it for gospel. Go take the Ford Falcon and tune the tranny. Good luck with 6th gear. I understand your math, the tranny, camber and other things are not what they seem. Example if you flip the tranny, maxing out final gear at 5000 and top speed at 124, you do not always get max Rpm and torque values. If you reduce final gear to 4725 you can get max RPM and torque. Look at some of Praiano's tunes he does the same thing. There is no 1 set formula that works for every car. It's PD doing their witchcraft.

 

[jblackrevo9]

I agree with all of you, there is a formula you can use. I got it from here


Correct me if I'm wrong, but it goes

1.399/x * 9100 = 7200

X = 1.768



But you have make sure your are setting the gears with in the powerband

 

[killerjimbag]

OK simple test go set up a NASCAR, give me your gear ratios. Don't forget first gear for initial excelleration, entering the pits etc, and the other gears for passing and drafting and your math goes out the window. You only have 4 gears should be simple.

 

[SiNS3V3N]

If your trying to reverse tune and your goal is to calculate the gears so that each shift lands at the exact same rpm, use this formula instead.

I made this formula to go from 6th to 1st.

It will start with 6th gear going backward, so you can tune 6th to hit top speed first then it will base calculations in reverse off that.

Base Data

Shift rpm = 7600

Desired entry RPM = 6500rpm

6th Gear = 0.902

Formula

6th Gear / (Entry rpm/shift rpm) = 5th gear

0.902 / (6500/7600) = 5th gear

0.902 / 0.8552= 5th gear

1.055 = 5th Gear

Rince, Wash, Repeat

This is To help with testing. I am NOT saying shifting into the exact same rpm is ideal. I'm just providing a formula to easily tune it to do so, this way people can test for themselves to see what happens when the gears all shift into the same rpm

With good equal spaced gears, each shift if shifting at the same rpm should land a little higher then the previous gear . Lowest entry rpm at 1-2 shift, highest at 5-6.

 

[killerjimbag]

If your trying to reverse tune and your goal is to calculate the gears so that each shift lands at the exact same rpm, use this formula instead.

I made this formula to go from 6th to 1st.

It will start with 6th gear going backward, so you can tune 6th to hit top speed first then it will base calculations in reverse off that.

Base Data

Shift rpm = 7600

Desired entry RPM = 6500rpm

6th Gear = 0.902

Formula

6th Gear / (Entry rpm/shift rpm) = 5th gear

0.902 / (6500/7600) = 5th gear

0.902 / 0.787879 = 5th gear

1.145 = 5th Gear

Rince, Wash, Repeat

This is To help with testing. I am NOT saying shifting into the exact same rpm is ideal. I'm just providing a formula to easily tune it to do so, this way people can test for themselves to see what happens when the gears all shift into the same rpm

With good equal spaced gears, each shift if shifting at the same rpm should land a little higher then the previous gear . Lowest entry rpm at 1-2 shift, highest at 5-6.[/quote[/quote
What about 4,5, speed tranny's ?]

 

[SiNS3V3N]

@killerjimbag

Them not hitting redline is understandable with "how" pd calculated speed, and how basic the graph readout is. The car will still rev up it's just the overly basic graph that falls short. It will also make the top of gears disappear if stretched to fast. That's not ill programming in the transmission, that's week programming of the graph readout.

PD is by no means perfect but the math here is rather basic and sound, just test any of my formulas and results will match perfectly, no variation at all. We do lack a 1/4 - full mile spot to test with accompanying data.

what about 4,5

Replace 6th with top gear.

Once done getting 5th, use 5th to get 4th, then 4th to get 3rd and so on.

Tune top gear to max speed or the GTP norm, top speed on longest straight lol. Then use Top gear to start the calculations.

 

[SiNS3V3N]

Test example

Stock RX-7 Spirit

Specs

8000 revlimit

Peak HP 292 @ 6500rpm

Peak tq 245 @ 4500rpm

Equation Data

Shift RPM 7600rpm

Entry RPM 5770rpm

Results

FDG) 4.300

Max Speed set to 193mph

1) 2.406
2) 1.827
3) 1.387
4) 1.053
5) 0.800
6) 0.607

FDG) 4.300

Instructions

If you shift @ 7600rpm every gear, then every gear will grab @ 5770rpm.

 

[killerjimbag]

Test example

Stock RX-7 Spirit

Specs

8000 revlimit

Peak HP 292 @ 6500rpm

Peak tq 245 @ 4500rpm

Equation Data

Shift RPM 7600rpm

Entry RPM 5770rpm

Results

FDG) 4.300

Max Speed set to 193mph

1) 2.406
2) 1.827
3) 1.387
4) 1.053
5) 0.800
6) 0.607

FDG) 4.300

Instructions

If you shift @ 7600rpm every gear, then every gear will grab @ 5770rpm.

I don't have any issues with my transmissions.

 

[SiNS3V3N]

Ideally the gears should be equal spaced to hit top speed. Then adjusted to be used on a given track most effectively, ie not shifting too often or too infrequent, eliminating unnecessary shifts if needed. Then using the tq curve (hp is based on this curve, tq is the direct source) we calculate our shift point to straddle the power trying to match power at exit as it falls, with power on entry as it climbs. Looking at the tq curve shows us the acceleration profile across the rev range, we want shifts to stay inside the rev range where acceleration is greatest. Engines that take high rpm to peak tq (Integra type-r peak tq @ 8300rpm) need really close gears to stay high in the revs to keep close to where acceleration is best. A V8 with tq topping off at 3500rpm hp running to 6500rpm will want gears falling much lower to stay in hard acceleration.

When it comes to matching power you take the tq curve and use it to calculate the geared power across the curve for each gear, as each gear multiplies tq, each gear will ultimately make different tq to the wheel.

You then have to cross reference the power at shift in the current gear to the power in the new gear at entry trying to equalize the 2.

The problem is the graph only shows peak figures with certainty, nothing else is clear so the curve pre peak tq and post peak hp are both mysteries. This make calculating shift points to power in this way impossible.

I don't have any issues with my transmissions.

Never said you did, this is a discussion that's been evolving. Nothing is directed at you, unless responding to any question you may post, if you post any. You asked a question, I answered. Take part or sit back and watch it's all good.

 

[SiNS3V3N]

Here is an example of what I'm talking about

Can't do this with only peak figures, we need a clear view of the curve.

This is the acceleration profile for 1st, 2nd & 3rd (example not based on the RX-7 example) so that people can get an idea about how to cross reference power in current gear vs power in the next gear. The math is not based on the power at 1 to 1, but each gears multiplied tq relative to the multiplied tq in the next gear. You do not try to match hp as if hp and tq are 1 to 1 in every gear, this is not the case, each gear is a power multiplier.

Torque x Gear

RX-7 torque for each gear @ Peak tq and peak hp.

Using RX7 gears from last example

Tq @ Peak tq----------tq @ peak hp

1) 590-----------574
2) 448-----------435
3) 340-----------331
4) 258-----------251
5) 196-----------191
6) 149-----------145

Only 4th gear is around 1 to 1

This is why 4th is used in a dyno pull, this is the gear closest to if not 1 to 1 in the gearbox, always the gear closest to 1 to 1 and NOT UNDER (overdrive) is used to do a dyno pull. This data can then be used to calculate the tq in the rest of the gears most effectively.

Right idea, just rough on the fine points. I've been doing this math since around 2000 when I just got started racing looking for every 0.0XX in the 1/4 mile. It's a language after a while.

 

[killerjimbag]

Ideally the gears should be equal spaced to hit top speed. Then adjusted to be used on a given track most effectively, ie not shifting too often or too infrequent, eliminating unnecessary shifts if needed. Then using the tq curve (hp is based on this curve, tq is the direct source) we calculate our shift point to straddle the power trying to match power at exit as it falls, with power on entry as it climbs. Looking at the tq curve shows us the acceleration profile across the rev range, we want shifts to stay inside the rev range where acceleration is greatest. Engines that take high rpm to peak tq (Integra type-r peak tq @ 8300rpm) need really close gears to stay high in the revs to keep close to where acceleration is best. A V8 with tq topping off at 3500rpm hp running to 6500rpm will want gears falling much lower to stay in hard acceleration.

When it comes to matching power you take the tq curve and use it to calculate the geared power across the curve for each gear, as each gear multiplies tq, each gear will ultimately make different tq to the wheel.

You then have to cross reference the power at shift in the current gear to the power in the new gear at entry trying to equalize the 2.

The problem is the graph only shows peak figures with certainty, nothing else is clear so the curve pre peak tq and post peak hp are both mysteries. This make calculating shift points to power in this way impossible.



Never said you did, this is a discussion that's been evolving. Nothing is directed at you, unless responding to any question you may post, if you post any. You asked a question, I answered. Take part or sit back and watch it's all good.

Do you use an automatic or manual transmission.

 

[killerjimbag]

Uhm check my profile. I'm up for discussion about many things.
Do you use a wheel?
Do you use an automatic or manual transmission? I noticed it says you race in real life. What do you race?

 

[SiNS3V3N]

Do you use an automatic or manual transmission.

Strange question as we are discussing optimum shift points. I feel a Red Neck "Here's ur Sign" joke coming. "No I drive automatic and make it shifts at the rpm I want with my Jedi force gear shift power." Kidding. Um Manually, wheel, absolutely no assist. When I was in my 20's started street racing until the law put a stop to it, then local track days with 2 1/4 mile tracks and 1 8th close by. ATM alimony, new wife, & 2 kids eat up my, trip to the track and repairs for trips to the track funds (pretty much if your not breaking stuff your not pushing hard enough) so until things in the economy get better I play GT6, and here we are.

 

[killerjimbag]

Strange question as we are discussing optimum shift points. I feel a Red Neck "Here's ur Sign" joke coming. "No I drive automatic and make it shifts at the rpm I want with my Jedi force gear shift power." Kidding. Um Manually, wheel, absolutely no assist.

Hardly a redneck. We should hook up and do some racing. Add my PSN

 

[Motor City Hami]

Hardly a redneck. We should hook up and do some racing. Add my PSN

Come on Killer... Dont you know Jack?

 

[SiNS3V3N]

Rebug PS3

Only offline as going on PSN gets a Console Ban, but I can do this

I will run TT offline on equal grounds, no hacks, I like good power and SS tires RWD

@Motor City Hami
That's not fair to say, he seems knowledgeable.

 

[killerjimbag]

Come on Killer... Dont you know Jack?

Silly me

 

[donpost]

Here is an example of what I'm talking about

Can't do this with only peak figures, we need a clear view of the curve.

This is the acceleration profile for 1st, 2nd & 3rd (example not based on the RX-7 example) so that people can get an idea about how to cross reference power in current gear vs power in the next gear. The math is not based on the power at 1 to 1, but each gears multiplied tq relative to the multiplied tq in the next gear. You do not try to match hp as if hp and tq are 1 to 1 in every gear, this is not the case, each gear is a power multiplier.

Torque x Gear

RX-7 torque for each gear @ Peak tq and peak hp.

Using RX7 gears from last example

Tq @ Peak tq----------tq @ peak hp

1) 590-----------574
2) 448-----------435
3) 340-----------331
4) 258-----------251
5) 196-----------191
6) 149-----------145

Only 4th gear is around 1 to 1

This is why 4th is used in a dyno pull, this is the gear closest to if not 1 to 1 in the gearbox, always the gear closest to 1 to 1 and NOT UNDER (overdrive) is used to do a dyno pull. This data can then be used to calculate the tq in the rest of the gears most effectively.

Right idea, just rough on the fine points. I've been doing this math since around 2000 when I just got started racing looking for every 0.0XX in the 1/4 mile. It's a language after a while.

But why bother drawing a torque curve for every gear and then working out the shift points, when you can use the single HP curve and get the same answer? Or you could if PD gave us more accurate graphs...

 

[axxxim]

If your trying to reverse tune and your goal is to calculate the gears so that each shift lands at the exact same rpm, use this formula instead.

I made this formula to go from 6th to 1st.

It will start with 6th gear going backward, so you can tune 6th to hit top speed first then it will base calculations in reverse off that.

Base Data

Shift rpm = 7600

Desired entry RPM = 6500rpm

6th Gear = 0.902

Formula

6th Gear / (Entry rpm/shift rpm) = 5th gear

0.902 / (6500/7600) = 5th gear

0.902 / 0.787879 = 5th gear

1.145 = 5th Gear

Rince, Wash, Repeat

This is To help with testing. I am NOT saying shifting into the exact same rpm is ideal. I'm just providing a formula to easily tune it to do so, this way people can test for themselves to see what happens when the gears all shift into the same rpm

With good equal spaced gears, each shift if shifting at the same rpm should land a little higher then the previous gear . Lowest entry rpm at 1-2 shift, highest at 5-6.

how did you get this number 0.787879
6500/7600 = 0.855263

 

[SiNS3V3N]

Typo from too much math with multiple examples and varying input data. The Equations is still sound.

Corrected

6th / (entry rpm / shift rpm) = 5th

0.902 / (6500/7600) = 5th

0.902 / 0.855263 = 5th

5th gear = 1.055

@donpost

Because power is not 1 to 1 in each gear, using the 1 to 1 power numbers from 4th in every gear is a miss representation of the power from one gear to the next.

If trying to match power in the current gear with entry power in the gear being shifted into we must know the ACTUAL power being made in each gear. The amount of power in each gear across the power curve is different for each gear based on how much the gear multiplies power, and how far apart the gears are set. How far apart they are set also changes the gap of actual power from one gear to the next.

Short answer, NO you cannot use the single hp curve to get the same answer, that's impossible and not how it's done. Each gear multiplies power differently, using the single hp curve is incorrect as no 2 gears make the same power this would be taking no account of how much power is ACTUALLY made in each gear.

We want to calculate using the actual power in each gear to compare. Only 1 gear has the numbers in the specs 1 to 1, none of the rest are the same.

To match power exiting one gear with how much power is made at entry of the next gear WE MUST know how much power is being made in each gear.

 

[DaBomm4]

If trying to match power in the current gear with entry power in the gear being shifted into we must know the ACTUAL power being made in each gear. The amount of power in each gear across the power curve is different for each gear based on how much the gear multiplies power, and how far apart the gears are set. How far apart they are set also changes the gap of actual power from one gear to the next.

Short answer, NO you cannot use the single hp curve to get the same answer, that's impossible and not how it's done. Each gear multiplies power differently, using the single hp curve is incorrect as no 2 gears make the same power this would be taking no account of how much power is ACTUALLY made in each gear.

We want to calculate using the actual power in each gear to compare. Only 1 gear has the numbers in the specs 1 to 1, none of the rest are the same.

To match power exiting one gear with how much power is made at entry of the next gear WE MUST know how much power is being made in each gear.

The gears do not make the power, they just translate it to the axles and then the wheels, the engine power never changes. Thus, only one graph for power should be needed.

 

[killerjimbag]

Typo from too much math with multiple examples and varying input data. The Equations is still sound.

Corrected

6th / (entry rpm / shift rpm) = 5th

0.902 / (6500/7600) = 5th

0.902 / 0.855263 = 5th

5th gear = 1.055

@donpost

Because power is not 1 to 1 in each gear, using the 1 to 1 power numbers from 4th in every gear is a miss representation of the power from one gear to the next.

If trying to match power in the current gear with entry power in the gear being shifted into we must know the ACTUAL power being made in each gear. The amount of power in each gear across the power curve is different for each gear based on how much the gear multiplies power, and how far apart the gears are set. How far apart they are set also changes the gap of actual power from one gear to the next.

Short answer, NO you cannot use the single hp curve to get the same answer, that's impossible and not how it's done. Each gear multiplies power differently, using the single hp curve is incorrect as no 2 gears make the same power this would be taking no account of how much power is ACTUALLY made in each gear.

We want to calculate using the actual power in each gear to compare. Only 1 gear has the numbers in the specs 1 to 1, none of the rest are the same.

To match power exiting one gear with how much power is made at entry of the next gear WE MUST know how much power is being made in each gear.

Which, according to your terrible math equations, you don't know. The last thing I think we need,is advice from someone who has a console/online ban, do to whatever they have done to deserve it. Your 15 minutes are up, move along.

 

[SiNS3V3N]

Look at the graph

Clearly it shows holding on to first until tq drops to 210 is ideal as 2nd doesn't start making more tq on the upside than 1st makes on it's down side of the tq curve.

@killerjimbag, no coffee this morning? "Terrible math" I don't know? I wrote the formula and it works don't get all upset when math is not subject to opinion it's right FACT

oh FYI Console is not Banned, going on PSN to play GT or COD with a JBPS3 (not many games do this) they flag your console ID for obviously being a JBPS3 then Sony Bans the Console ID. Most games don't care play online it's all good. I stay away from all games just in case to keep access to PSN for DLC the more you know huh..

If you like it or not I speak facts not Theories, sorry "Mixed Theories" based on half truths and half understandings of truths.

No need to be a negative Nancy. The facts are in front of you, ignore them and stay blind or try to understand and maybe get enlightened.

 

[DaBomm4]

The translation only magnifies the power, it does not increase the power. Thus, still, multiple graphs are unnecessary.

 

[SiNS3V3N]

The translation only magnifies the power, it does not increase the power. Thus, still, multiple graphs are unnecessary.

Semantics

Multiplying power IS power magnified.

Same thing

Po-(tay)-to Po-(tah)-to

 

[DaBomm4]

It is still the same base power, just used differently, the optimal point is the same, just there is no need for more than one graph for it.

 

[SiNS3V3N]

It is still the same base power, just used differently, the optimal point is the same, just there is no need for more than one graph for it.

Semantics


We are supposed to be looking at how "the base power" is used differently gear to gear and thus we multiply the figures respectively. Only 4th is 1 to 1 (like a record baby)

Your point moot

Then saying we only need one graph is actually wrong, we use multiply graphs to compare power magnified in each gear respectively.



Again

Po-(Tay)-to / Po-(Tah)-to

Your playing semantics, that is unnecessary and only serves to confused instead of clarify.

 

[DaBomm4]

There is no need for more than one graph, the power is the same.