Support Board
Date/Time: Thu, 06 Feb 2025 02:04:26 +0000
Sierra chart is very laggy after update
View Count: 2340
| [2024-12-04 15:05:18] |
| frozentundra - Posts: 93 |
|
Updated to 2716 this morning and everything is running very fast. Without a doubt a performance increase over 2710. My global chart update interval = 1000ms. My most demanding chart I set to update interval of 100ms and it is experiencing zero lag. Thank you very much! |
| [2024-12-04 15:22:56] |
| User753428 - Posts: 167 |
|
Makes completely no sense at all. Not at all reproducible by us.
We are using settings of 10 ms global Chart Update Interval: General Settings Window: Chart Update Interval (Global Settings >> General Settings >> General >> Update Intervals) A multiplier setting of 1: General Settings Window: Timer Management - Chart Calculation/Draw Time Multiplier for Interval Delay (Global Settings >> General Settings >> General >> Update Intervals) And a maximum delay of 1000: General Settings Window: Timer Management - Maximum Interval Delay in Milliseconds (Global Settings >> General Settings >> General >> Update Intervals) We have about 60 charts open. Sierra Chart is completely responsive. No issues. Fast updating. Everything works fine. No trace of any problem at all. Performance is excellent. All of this is fact. when you were testing this with global Chart Update Interval set to 10ms, may i ask what your Remote Buffer Delay Send Time was set to? |
| [2024-12-04 15:24:56] |
| Sierra_Chart Engineering - Posts: 18297 |
|
20 ms. That is the lowest possible. But that has no effect on Sierra Chart performance on the client side. In other words it does not change, calculation or processing times.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, use the Teton service: Sierra Chart Teton Futures Order Routing |
| [2024-12-04 15:41:36] |
| Sawtooth - Posts: 4169 |
|
In v2716, if I set the Chart Update Interval to 15000ms, there is no stalling of rapid cursor movement. Is this too large of a setting for any other reasons? |
| [2024-12-04 16:24:41] |
| Sierra_Chart Engineering - Posts: 18297 |
|
No problem with that setting. Whatever you want to use. It just means that for charts that are relying on that particular setting they are only going to update every 15 seconds.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, use the Teton service: Sierra Chart Teton Futures Order Routing |
| [2024-12-04 17:24:28] |
| Sawtooth - Posts: 4169 |
|
It just means that for charts that are relying on that particular setting they are only going to update every 15 seconds
How is this an improvement?
|
| [2024-12-04 18:50:31] |
| Rui S - Posts: 196 |
|
Just my contribution: Due to the feedback from several other users regarding these issues, I have been postponing to update my platforms. Today I had some free time so I decided to make a test and updated one of my platforms to v2716. Then, I started testing and opened the very same chartbook in two separate instances (with my default settings) side by side, one with the upgrade to v2716 and the other with my current v2700. The immediate result I noticed was that with v2716 the CPU load was much lower than with v2700. I got really puzzled with it because it goes against any report made by every other users. Then I realized that there was a lag on v2716 when comparing the price movement (charts / DOMs side by side) with v2700 (both connected to the same SC server). So, I went through the new settings on v2716 one by one and made some experiments. To make this report shorter, in order to get a very similar performance with the two instances, I have to use the following settings: - Timer Management - Chart Calculation/Draw Time Multiplier for Interval Delay = 0.5 - Timer Management - Maximum Interval Delay in Milliseconds = 250 - Trade DOM - Had to reduce from 50ms to 25ms (half) - Main charts - Had to reduce from 200ms to 100ms With this settings, I get a very similar performance out of my main two chartbooks, whereas the CPU load becomes very similar in both instances. Just to make it clear, I tested twice with two different chartbooks (one at a time) and both with the "twin" charts and DOMs side by side. I haven't experienced any lag with the pointer and besides having to alter the mentioned settings, everything seems normal. To finish, I think it would really be a very good idea to have the High, Medium or Low priority option for the charts as mentioned by SC Engineering Date Time Of Last Edit: 2024-12-04 18:53:36
|
| [2024-12-04 18:59:23] |
| User925794 - Posts: 21 |
|
Regarding v2716 - it runs my chartbooks now, but still laggy in every aspect. Most concerning is the lag when using the cursor and the trade window on my chart - e.g. click on BUY BID, wait for it....ok, order applied....this never happened in v2700 and before. The cursor and pull down menu action are also still laggy. Another concern for me, I use the Spreadsheet for Trading to exit trades at target and I'm more than a little suspicious of that response time. FYI, I did go thru my charts and any chart that has bar times >60min got a setting of 30000 ms. The others are at 0 = Global which is set @500. I see others are trying to "tune" by what seems like a random "try this" approach. IMO, A more logical approach is needed. I get it, windows is not my daily driver either (Linux user here). I only use Windows as an appliance OS for running SC. I've tried to look up MS timers and came across some gamers that have various hacks to these little demons to improve their FPS experience and of high suspect is the timer HPET so I disabled that one and it didn't seem to make a difference in any release. Anyway, since Team SC can't seem to reproduce any of the ill effects many of us are having, perhaps it's a HW to OS specific problem as HW OEM's do a lot of work when it comes to drivers and getting MS stable on their HW releases? Grasping at straws here because in order to trade live, I need at least v2700 response times (near real time). Another problematic test case was the 2 users who rebuilt their chartbooks/charts which solved their problem - I agree - how the hell could that solve the problem as it doesn't make sense, but there it is - and oh how I dislike the thought of doing that, but I will if that's what it takes. Going back to v2700 until I can't. |
| [2024-12-04 19:20:14] |
| User753428 - Posts: 167 |
|
Today I had some free time so I decided to make a test and updated one of my platforms to v2716.
Then, I started testing and opened the very same chartbook in two separate instances (with my default settings) side by side, one with the upgrade to v2716 and the other with my current v2700. The immediate result I noticed was that with v2716 the CPU load was much lower than with v2700. I got really puzzled with it because it goes against any report made by every other users. Then I realized that there was a lag on v2716 when comparing the price movement (charts / DOMs side by side) with v2700 (both connected to the same SC server). So, I went through the new settings on v2716 one by one and made some experiments. To make this report shorter, in order to get a very similar performance with the two instances, I have to use the following settings: - Timer Management - Chart Calculation/Draw Time Multiplier for Interval Delay = 0.5 - Timer Management - Maximum Interval Delay in Milliseconds = 250 - Trade DOM - Had to reduce from 50ms to 25ms (half) - Main charts - Had to reduce from 200ms to 100ms With this settings, I get a very similar performance out of my main two chartbooks, whereas the CPU load becomes very similar in both instances. Just to make it clear, I tested twice with two different chartbooks (one at a time) and both with the "twin" charts and DOMs side by side. I haven't experienced any lag with the pointer and besides having to alter the mentioned settings, everything seems normal. this is similar to my experience. i haven't experienced any of the UI lag experienced by others in the newer versions (this may be because i have very simple charts) but the price updates (chart update interval as perceived by the human eye) is for sure slower on the newer versions. i've been using 40ms chart update interval + 40ms Remote Buffer Delay Send Time for years now with no issue. this setting is perfect for both my trading needs and what my computer/internet is capable of from my location. but the problem i run into when using Rui S.'s solution for this is that when i reduce the chart update interval + remote buffer delay send time in 2716 by half (so 20ms + 20ms), i start running into increased transmissiondelay due to more # of outstanding send buffers. so 40ms/40ms in v.2700 isn't the same as 20ms/20ms in v.2716. i tried 20ms chart update interval + 40ms Remote Buffer Delay Send Time in v.2716 but when i do that, unfortunately price updates are visibly slower than at 40ms/40ms in v.2700. |
To post a message in this thread, you need to log in with your Sierra Chart account: