Any other variable (be it FOV, Steam branch, game, etc. It is a PiTool or Firmware bug, and can be reproduced 100% by selecting 65Hz, and immediately vanishes when switching to 90Hz or 120Hz. The minimal smart smoothing artifacts are identical in 60Hz/30fps as 75Hz/37,5fps.įYI it is present in the most actual PiTool 254 and FW 255 for the 5K+ SN204. X-Plane in 60Hz and Smart Smoothing at 30fps runs smooth basically everywhere on my system. Some good news is also Im testing a 60Hz firmware on the 8KX and so far it works perfectly. I will first do a round-up of the performance in 6 or 7 flight simulators including X-Plane 11, and then in upcoming videos I will dig deeper into the settings and optimization. PiTool 1.0, SS 70-72% (due to Parallell Projection, resolution is set to as close to 3800pix horizontally as possible) The settings Ive found looks best and still being “doable” are:
![x plane 10 vs 11 performance x plane 10 vs 11 performance](http://www.x-plane.com/wp-content/uploads/2017/05/X-Plane_11_DVD_cover.png)
Either way, with lots of tweaking (Process Lasso, FlyWithLua NG, XP11SettingsTool, overclocking etc) a stable fps mostly above 40 its really not bad - and yeah it looks awesome! Running with Smart Smoothing enabled (37,5fps) makes the flights totally smooth as well, with very minor artefacts.
X PLANE 10 VS 11 PERFORMANCE FULL
Remember, these sceneries are extreme when it comes to world objects, textures and mesh.Īs you know, X-Plane11 tanks the CPU and GPU because of the bad optiomization, and there is no way to run this thing in a full 75fps frame rate. World Objects is kept on Medium because everything above kills the CPU and lowers the framerate by 10-15fps or more. Basically as intense at it gets when it comes to scenery. ORBX TrueEarth US Northern California (around SF), ORBX LOWI Innsbruck, ORBX TrueEarth Washington and others. Im using ORBX scenery to really show the most GPU/CPU intense scenarios. This time I have chosen to really take the time and find an optimal balance between image quality and acceptable performance. Just want to give you a hint about the performance here. PostgreSQL 13 represents another leap forward for Open Source relational databases, as it continues to improve with every new release.Since a few days ago I’ve been messing around a lot with X-Plane 11 VR with the 8KX for some upcoming videos (including other VR flight sims). Version 13.0 performs better than 12.4 out of the box, and the difference is even more striking when both servers are tuned. The tests show that there is immense benefit to properly tuning your PostgreSQL server. If we compare 12.4 and 13.0 with proper tuning, we can see a near 20% improvement in transactions per minute at 250 virtual users before slowly dropping back down to 12.4 levels. The same can be seen between 13.0 untuned and tuned. When properly tuned, we can achieve up to eight times the number of transactions per minute, and sustain that as the number of virtual users increases.
![x plane 10 vs 11 performance x plane 10 vs 11 performance](https://oceanofgames.live/wp-content/uploads/2019/07/x-plane-11.png)
We can see from this chart that the untuned performance of 12.4 and 13.0 are mostly identical, with 13.0 having the edge, and performance maxes out between 50 and 75 virtual users. The values indicate transactions per minute. The following table shows the results for 12.4 and 13.0 both out of the box and properly tuned. Note: for a proper TPC-C benchmark, the test must be run for a minimum of two hours. This does not reflect a real-world workload, nor is it appropriate if running a "by the book" TPC-C benchmark. Note that think and key times are disabled, as the intention with this test is to see the maximum possible performance. HammerDB is then run with a ramp up time of 5 minutes and test time of 1 hour.
![x plane 10 vs 11 performance x plane 10 vs 11 performance](https://store.x-plane.org/assets/images/files/alabeo/c207xp11/c207_skywagon_12.jpg)
A base backup of this pristine database is preserved.įor each number of virtual users, a new database is restored from the base backup. ANALYZE and VACUUM FREEZE are run on all tables. Each table is CLUSTERed by its primary key index (except for history which has no indexes). The database is populated by HammerDB with 2000 warehouses. No changes were made to the algorithms used by HammerDB. These modifications are in accordance with the TPC-C specification:įor each table, the following list of attributes can be implemented in any order, using any physical representation available from the tested system.
X PLANE 10 VS 11 PERFORMANCE DRIVER
The test driver used is HammerDB with a few modifications to the data model to use appropriate data types for PostgreSQL, optimize column order for alignment, and remove fill factor settings (see the Data Model section at the end of this post for the exact schema used). All other settings remained at the default values.