I have been a Matlab loyalist for 25 years
I have been a loyal MATLAB user for 25 years, starting from my university days. While many of my peers migrated to Python, I stayed for the stability, compatibility, and clean environment. However, I am finding the 2025 version exceptionally laggy. Despite running it on an $10k high-end machine, simple tasks like viewing variables and plotting take up to 60 seconds - actions that were near instantaneous in the 2020 version. I want to stay continue with MATLAB, but this performance gap is a major hurdle and irritation. I hope these optimization issues can be addressed quickly.
22 件のコメント
新しい順
I have been near-daily using Matlab since it's very first version in the 80's. R2025b is the worst. Very laggy and doesn't retain some of the better features of the past few versions. I've bought personal licenses occasionally over the decades for hobby use at home and my company (in Mathworks's backyard) also has a multi-decade corporate license for 200+ users. We recently got R2026a and I will install it. If that doesn't go over well with me, I'm going to follow many of my peers and ditch Matlab for the Python environment. So sad.
R2026a is an improvement over R2025b, but a lot of work is still needed. My tolerance for these versions is a byproduct of just how bad things had gotten for Mac users in the 2022-2024 era, where certain graphic operations would randomly (and frequently) crash MATLAB. Some people claim that changing the Java distribution resolved the problem, but that never worked for me. This forced me over to the JavaScript side of things, i.e. using uifigures in place of figures, though now that distinction has largely disappeared.
Given that the Java to JavaScript graphics transition was started over ten years ago, it's disappointing how slow and buggy things are.
Very much similar experice here. Updated from 2023b to 2025b. Before, matlab updates have always been rather effortless with improvements. The 2023->2025 experience was a not so much. The new ML environment seems littered with issues. Littered with bugs and performance issues.
I have been using matlab for over 15 years with many different version. Over that time i have been building tools that make building more complex tools easier. With each version of matlab there would be some things that i had to change to be compatible but they were minor. I have been testing r2025 and now recently r2026a and many of the tools that i have created over the years are unusable. I think there are not only key problems with some of their functions but also I dont think the drawnow actaully works anymore. I think it does spend time within the drawnow but i do not actually see that it does anything.
I have several performance issues that all circle around the debugger and variable editor and only occur after having used the debugger for a longer session:
- If a variable is already open, CTRL+D does not move the editor-highlighted variable to the foreground
- After some time, the variable window freezes and needs to be reopened
- Variables are no longer shown (although whos shows them) if one moves around the stack
- moving between stack depths takes a longer time
- deeper stacks are way slower than on earler matlab versions (especially when running tests, that add at least 10 function calls to each stack)
- Warnings and printouts are very very slow in the new web-based command window. Sometimes warnings contain lists and printing that to log takes a longer time compared to earlier matlab versions
- when the debugger stops, the variable view moves multiple variables into the foreground
Moreover, the memory usage is increased and the same workloads need twice the amount of memory on hpc than in 2024b (in batch usage, so desktop issues should not apply here)
In short: I am extremely worried about the direction that mathworks is heading. Releasing desktop versions prematurely and sacrificing one of the few advantages that matlab has over other languages: a very simple and fast debugger that works flawlessly. I do not think that is a wise strategy.
I’ve also been struggling with performance issues lately. The basic MATLAB environment is often unbearably slow (tens of seconds), and Simulink is even worse. I really hope MathWorks addresses this—maybe it’s time to rethink the underlying framework.
---------2026.7 update----------------
Why I'll Keep Using MATLAB Long-Term (16+ Years of Experience)
I've been using MATLAB continuously for over 16 years. Many people complain about its performance bloat — and those issues are indeed real. But on the flip side, MATLAB was never meant to be a "general-purpose" language. Striking the right balance between software usability and programming language flexibility is extremely challenging, especially when you factor in cross-platform support and OS compatibility.
It's 2026 now, and I will continue to stick with MATLAB for the foreseeable future. Here's why:
- Personal learning speed can never keep pace with the rate of technological change. MATLAB consistently distills fast-moving technologies into built-in functionality, allowing us to quickly get up to speed in any given field. This saves us from having to invest significant time learning a specialized third-party toolkit from scratch.
- MATLAB captures the core direction of every cutting-edge field. The key updates across its various toolboxes keep you relatively ahead of the curve, without having to piece together fragmented trends on your own.
- Backward compatibility is outstanding. Code written 20+ years ago runs on modern development environments with little to no modification. When incompatibilities do arise, only minor tweaks are needed.
- MATLAB is essentially a "grand encyclopedia" of scientific computing. Whenever I encounter something unfamiliar or want to explore a new domain, the first thing I do is check the official documentation.
- Installation and configuration are straightforward. Compared to general-purpose programming languages, where you have to worry about platform, version, and IDE differences, MATLAB lets you install the toolboxes you need with one click. The integrated environment is clean, intuitive, and ready to use right out of the box.
"MATLAB was never meant to be a "general-purpose" language. Striking the right balance between software usability and programming language flexibility is extremely challenging, especially when you factor in cross-platform support and OS compatibility."
The creeping take-over by enterprise programming concepts and the attempt to turn MATLAB into a general-purpose language will be its downfall. There are plenty of very good general-purpose languages out there (hello Python!), but there are few languages that treat mathematics as a first class principle.
When I ask people what they like about using MATLAB it often involves the IDE, the easy graphics, the debugging integration, etc. When these get left behind, then users are getting left behind.
I recently downgraded to 2026A from 2024B, feels like I stepped on a few 1000 rakes.
@MathWorks Support Team Remember downfalls start very slowly then all at once!
If you’ve ever had an equation that gives you heartburn — Colebrook, β–θ–M, Prandtl–Meyer, Rachford–Rice, or any other implicit monster — drop an image of it here. I’ll convert it into strict algebraic format and run it through a deterministic solver I built that isolates the unknown without rearranging, iteration, or guessing. You’ll get the solved value back with 9–14 decimal‑place precision.
@Jan Christian Could I contact you please to arrage a video call? I've reported your case to development and they need a little more information.
I understand you @Jan Christian. I've also been a MATLAB user for many years and have always been obssessed with MATLAB performance. Making MATLAB code faster was the backbone of how I started my career as a Research Software Engineer.
Since I've joined MathWorks, performance is still once of the top things I concern myself with and MathWorks internal enhancement tracking database is full of issues from me saying things like 'I've noticed this could be faster' or 'I saw this user complaint on MATLAB answers and reproduced it'. Even though I am not in development, one of the things I'm proud of is that there are always several things in every MATLAB release that I can point to and say 'The internal thread that led to that enhancement came from me' and many of them are performance related. There are many at MathWorks who do similar.
I'm telling you this to assure you that comments like yours are noticed, read and acted on. Thank you for sharing and please keep the feedback coming.
This thread led me to including a performance section in my blog post about R2026a MATLAB R2026a has been released – What’s new? » The MATLAB Blog - MATLAB & Simulink. In the blog I only cover a fraction of what's been made faster -- the full list is in the release notes.
The performance that I was talking about in another comment of this thread, and what I believe the original poster intended, is macroscopic. Specific functions are very easy to clearly define and measure. However, overall use of the app is often harder to diagnose.
Unfortunately, I am in the same camp as @Jan Christian in that I have a very expensive computer (Intel Mac) on which R2025a and R2025b are very laggy compared to R2024b. Specific slow-down scenarios include stepping into and out of the debugger and opening the variable editor. While I have no doubt that much of the built-in functionality of R2025b is faster than that in R2024b, on some hardware other aspects of the app have become so much slower as to make all such performance improvements unnoticeable.
The debugger has definately slower post-Java. One of my new pet peeves is how a debugger tab gets added to my minimal "command window only" interface. This happens whenever time debugger starts, i.e. I have to close that tab every time. Maybe I'm a purist, but screen space is still precious to me...
I can confirm slowdowns on the Mac side, where I use MATLAB most often. I had to abandon earlier releases becuase they repeatedly crashed not matter what Java library I tried.
サインインして参加する
タグ
