2025 Scuderia Ferrari F1 Team

[atanatizante]

Some suggest it is an “escape,” given the unresolved issues of Ferrari’s V6 turbo, which has pushed extreme boundaries of combustion pressures and temperatures, requiring interventions on cylinder heads to avoid reliability failures.

This may be why Mercedes engines are seen as the benchmark, with Ferrari’s unit rated slightly lower.
https://it.motorsport.com/f1/news/f1-fe ... /10758225/

[venkyhere]

Outdoing computer expectations is not easy. And seeing a 12° delta lower on the tire barely trying via a tiny change!
I bet they understood, you don't become a 7-time world champion by luck :

Unless the data on F1insightshub.com is wrong, I can't see how this guy's YT video is accurate at all. Of course there is no steering trace on this website, but things like throttle, brake and speed traces point the exact opposite of what that video suggests. I specifically looked at lap27, and can see that the claim in that video is wrong. Then looked at lap26 and lap28 and it's the same story there as well. Then I took lap5 where LEC was slower than HAM, as he had finished off his battery in the fight with Piastri and was on recharge without DRS, while HAM had DRS and was 0.5s faster than his teammate. Even there, HAM is losing time to his teammate through the Lesmos. I tried finding a random lap where HAM was faster than LEC through Lesmo2 (as the video claims) and gave up after a few attempts, because that wasn't the case. If the guy who posted the video can actually come up with the data trace that shows that what he is alluding to is true, then I will believe that video.

LEFT : lap27 (the one claimed by the video) :
and
RIGHT : lap5 (the one where HAM's lap was 0.5s faster than LEC's) :
-- open image in new tab for better readability ---

[PlatinumZealot]

Outdoing computer expectations is not easy. And seeing a 12° delta lower on the tire barely trying via a tiny change!
I bet they understood, you don't become a 7-time world champion by luck :

Unless the data on F1insightshub.com is wrong, I can't see how this guy's YT video is accurate at all. Of course there is no steering trace on this website, but things like throttle, brake and speed traces point the exact opposite of what that video suggests. I specifically looked at lap27, and can see that the claim in that video is wrong. Then looked at lap26 and lap28 and it's the same story there as well. Then I took lap5 where LEC was slower than HAM, as he had finished off his battery in the fight with Piastri and was on recharge without DRS, while HAM had DRS and was 0.5s faster than his teammate. Even there, HAM is losing time to his teammate through the Lesmos. I tried finding a random lap where HAM was faster than LEC through Lesmo2 (as the video claims) and gave up after a few attempts, because that wasn't the case. If the guy who posted the video can actually come up with the data trace that shows that what he is alluding to is true, then I will believe that video.

LEFT : lap27 (the one claimed by the video) :
and
RIGHT : lap5 (the one where HAM's lap was 0.5s faster than LEC's) :
-- open image in new tab for better readability ---

https://i.ibb.co/mF6PLPY4/Hamilton-Monz ... -video.png

What is the source of those traces?
In recent years I have seen people publishing these and I can only guess they are compiled from screen grabs of the onboards? Otherwise does the FIA publish the telemetry?

[venkyhere]

What is the source of those traces?
In recent years I have seen people publishing these and I can only guess they are compiled from screen grabs of the onboards? Otherwise does the FIA publish the telemetry?

F1 releases telemetry data (don't exactly know who, whether it's FoM or FIA or the teams themselves) via python packages like FastF1 or OpenF1 (the ones I know of) ; which, would then be accessed by the public using Python scripts to build 'analysis' and then use various database/math tools to publish graphs, plots etc on their websites. The one I refer to, is F1insightshub.com which our member Emag kindly hosts. There are others like f1tempo.com, racingstatisticsf1.com etc. My posts are nothing but screengrabs from these websites.

It's definitely not 'screengrab' from telecast (what do you anyway see on the telecast, other than laptimes and at best sector times?)

[Luscion]

Don't know how reliable, but apparently Zimmerman and Schmidt have already completed the conceptual design of the power unit and tests are being scheduled for the Dyno. The 2026 power unit has been optimized in terms of its performance at higher internal temperatures, with less heat exchange.

Also this,

"In close synergy with the aerodynamics department, reinforced by the arrival of Franck Sanchez, it is believed that the internal fluid dynamics of the 2026 Ferrari will be particularly innovative, based on a strong boost from the power unit's packaging."


They also talk about a rumour that was spread through the paddock that merc engines would be well ahead and that Ferrari were struggling, but that's it'sjust a political game to sway engineers not to move to Maranello, merc have denied people from their PU department went to Ferrari but their sources say otherwise , significant technical transfers between the top 3 teams expected


https://racingnews365.com/surprise-ferr ... power-play

[dans79]

F1 releases telemetry data (don't exactly know who, whether it's FoM or FIA or the teams themselves) via python packages like FastF1 or OpenF1 (the ones I know of) ; which, would then be accessed by the public using Python scripts to build 'analysis' and then use various database/math tools to publish graphs, plots etc on their websites.

They aren't releasing it, people are pulling if from the api that's feeds the app. Several years back we had threads about it, such as this one for example.
viewtopic.php?t=28804


What no one knows, is how good the data actually is, because everyone involved has motive not to provide high quality data. Thus the data has likely been massaged in several ways to make it look good, and potentially be informative at a high level, but no ones is going to get detailed information from it.

[Emag]

F1 is technically releasing the data, but not explicitly for the purpose of building telemetry analysis tool. They're just made available for use on their own services (such as F1TV).

The real source of the data is the livetiming service F1 has developed. What happens in that service internally, we don't really know (as far as I am aware), but the data that we use to display telemetry comes from the livetiming endpoint :

base_url = 'https://livetiming.formula1.com'

Now technically speaking, this data is supposed to be available to those who have a livetiming subscription through F1, but it was unprotected for a long time. Recently they changed it so that you cannot fetch live data (as in while a session is ongoing), but the /static/ endpoint is still free to use and that's where the data is coming from. People who have developed these telemetry services (like me) and made them public are technically in F1's hands. At the moment, we are not fetching data that F1 makes public anywhere else in any of their services. But if F1 decides to write their own Telemetry Analysis platforms at some point (for money), then we all will be forced to shut the public things down.

For the time being though, it's available for use. You can try it yourself (here is the [compressed] telemetry data of the 2023 Bahrain Grand Prix Race for example) : https://livetiming.formula1.com/static/ ... jsonStream

What FastF1 (and other libraries) do, is that they get this compressed data, process it, and make it available to use on easy-to-work-with data structures.

FastF1 is a really well-developed and extensive API, and there's a lot going there, but their api service specifically (the one responsible for fetching the data), can be checked here

So what is the data exactly? In short, it is what F1 uses to display the HUD on F1TV or on their web livetiming service and it is actually coming directly from the transponders in the cars. This means that data quality-wise, this is definitely not some mumbo-jumbo useless crap, but unfortunately it's prone to errors and inconsistencies (I guess people can kind of tell considering how often TV graphics glitch out). The data is often sampled at different frequencies on different laps, recording of data often starts missaligned for different laps. There can be data corruption during specific point in the lap. DateTime on specific sample points may be off and many more possible issues ...

Some of those issues, FastF1 deals with internally. Others are a little bit harder to deal with. F1Tempo (as far as I am aware), uses FastF1 functions out of the box, so the telemetry there can vary a lot lap to lap. The new one (GPTempo), seems to do some processing before displaying it to the user so it's better than their old site. If you prefer their interface, I would recommend using GPTempo instead of F1Tempo.

As for me personally, I also do a lot of post-processing to the data that comes from FastF1 and you will notice in F1InsightsHub there's rarely weird delta spikes in the braking zones and whatnot. Also, I try to perform alignment so that sector gates match exactly to the driven laps. Quality of alignment of course depends a lot on the quality of the original data, so that's why I put in the Delta Confidence to let people know when source quality was bad and I had to extrapolate to perform the alignment.

[ricardo91]

A very good in depth video covering the rumoured changes Ferrari will make leading into 2026 on the engine side. While we always hear these sorts of rumors every year it's an interesting listen. It showcases the Direct Metal Laser Sintering they could be using to develop the engine. I have no idea if it's feasible but here you go!

Also the engine is probably not as bad as believed - looks like Wolff has been busy spreading rumors inside the paddock. Let the mind games begin

[PlatinumZealot]

What is the source of those traces?
In recent years I have seen people publishing these and I can only guess they are compiled from screen grabs of the onboards? Otherwise does the FIA publish the telemetry?

F1 releases telemetry data (don't exactly know who, whether it's FoM or FIA or the teams themselves) via python packages like FastF1 or OpenF1 (the ones I know of) ; which, would then be accessed by the public using Python scripts to build 'analysis' and then use various database/math tools to publish graphs, plots etc on their websites. The one I refer to, is F1insightshub.com which our member Emag kindly hosts. There are others like f1tempo.com, racingstatisticsf1.com etc. My posts are nothing but screengrabs from these websites.

It's definitely not 'screengrab' from telecast (what do you anyway see on the telecast, other than laptimes and at best sector times?)

The telemetry is very low resolution so that indicates to me it is a "grab" or purposely downgraded or data from a reduced table.

They used to have software that grabbed information from the onboards. RPM, throttle, braking, Kers, steering angle, speed. It was possible when onboards became available.

If thete is a link to the tables or datasets released by the FIA or FOM, as the premier f1 technical site I think we should make a thread featuring them.

[Sphere3758]

Getting onto Baku : I remember Carlos Sainz talking about how Charles is able to get more from his car in Baku because of his ability to ride lower i.e. handle the bumpiness of the track better than he could. Given that there are no real high speed corners on this track and that Charles can probably ride lower than others, is there a chance ?