ThinkPad P-Series: Portable Workstations with NVIDIA Quadro RTX5000 (ThinkPad P53)
ฝัง
- เผยแพร่เมื่อ 29 ก.ย. 2024
- In this video I look at a high-end Lenovo ThinkPad P53 laptop equipped with a NVIDIA Quadro RTX5000, 32GB Ram and an Intel I7 with 12 threads. This computer comes preinstalled with the Ubuntu operating system, NVIDIA drivers, and the NVIDIA data science distro/pipeline. I will use this computer for a series of videos looking at Linux as a desktop operating system for data science.
Lenovo ThinkPad P-Series:
www.lenovo.com...
Lenovo and Linux:
techtoday.leno...
news.lenovo.co...
news.lenovo.co...
NVIDIA Data Science Stack:
github.com/NVI...
![ได้เวลา! พาน้องแฝดฉีดวัคซีนครบ 1 เดือน🫣 [cc] แดนแพทตี้ SS2 | EP.52 |](http://i.ytimg.com/vi/IzeLtaXBLNE/mqdefault.jpg)
Plot Twist: Jeff installs Windows to test Flight Sim ✈️
Is it to heavy to be as my main laptop to travel amd go out?
I have a real dislike to setting up laptops with nvidia gpus along with integrated gpu. Just takes way too much work. Cuda installs, cudnn install, but tf doesnt see it, then try again in a different repository...... driver version -> cuda -> cudnn doesnt match, but I specified the versions...ahhhhhhh ... If Lenovo is doing the work, I would definitely pay for it. I run rtx 5000 in my desktop would like to see how much slower the rtx 5000 in laptop form
I migrated from the T-series to the P-series because of the P53, despite the T-series being better for use when traveling or in places other than your home or office. However, if you don't need to move the laptop and it will remain stationary, the P-series is perfect.
my p53 is the best thing i have ever gotten. ive dropped it down flights of stairs with barely a scratch on it! dual booting was a breeze and it works way better than i could have dreamed!
I definitely like how the RAM and battery are not soldered to the mobo; that allows upgrading and repair options, which is a big plus.
But is the Quadro series GPU a worthwhile card for our purposes? The RTX series cards are much cheaper, from my understanding and nearly as capable for deep learning.
Are there any Lenovo laptops that have an RTX card instead of a Quadro? This should make the laptop much more cost-effective.
Also, great best practice using Docker for your virtual environment setup. I got so tired of fighting Nvidia drivers in a Conda environment; Docker is a big help for that.
It is a NVIDIA® Quadro® RTX™ 5000 16GB. Should have clarified that! Thanks.
NVIDIA Quadro RTX5000 This video card has Fermi Architecture. I think it is outdated for using many python libraries like RAPIDS, etc.
no one:
Prof. Heaton: flexing with RGB keyboard
xD love your videos
Hah, yes RGB, maybe a guilty pleasure. And it does look cool on some of my lenses that can do the blurry background thing.
What you prefer p53 or P15 with same spec?
I rather like my Lenovo P73 i9 128GB RAM 6TB hard disk "laptop"... except that it gets rather hot.
Quadro 5000 is good but most of us cannot afford this specs... i was searching for different laptops last month for doing some object detection project.. i tred razer blade 15 and Alienware 15 R3 and finally end up keep razer... i feel like RTX is better for deep learning than Quadro which is good for CAD related processing. but Thinkpad has the best keyboard for sure..
Would be interesting to see if that famous throttling issue of Lenovo is still present.
I will look for that when I push the CPU the next time.
@@HeatonResearch any updates?
Hi, i tried to install fedora on my Quadro RTX5000 (ThinkPad P53) but it wouldnt what could be the problem. only ubuntu can be installed on it . Thanks
Thanks for the info. I usually use Ubuntu, myself.
Hey Jeff, comments and questions.
First the question... Is a Quadra card a good GPU for Deep Learning? I understood it was more aimed at Professionals in the CAD/CAM and High-End 3D modelling fields and not so much the Machine Learning/Data Science fields?
Comments. I love my Lenovo X1 Extreme. It’s GPU is pretty wimpy but that doesn’t bother me because it’s my primary development/testing machine and even if I had a huge gpu in it I wouldn’t want to train a large model and basically lose my computer for hours, days or even weeks. Having a gpu in the laptop is awesome as a development force multiplier but I still rely on (currently) cloud GPUs for any real model building. As far as Lenovo’s higher ended machines one of their best features is the keyboard. I’m more programmer than scientist/researcher, I guess I’m a deep learning developer and spending hours on the X1 keyboard is only possible due to how great it is. Looking forward to the rest of the series and other content. It’s always enjoyable and more often than not I learn at least 1 thing and often more in the vast majority of your videos so keep it coming. I’ve pointed pretty much anyone who asks what is a great beginners channel to watch and yours always comes to mind in my very top recommendations. Another nice thing about the X1 is it’s as light and thin as any high end machine can be. This is a trade off. I went this route because I follow the development workflow above but even still at times I really wish I had a heavier device like a System76 Oryx Pro with a real gpu when there are model generation tasks that are small enough to build in a timely manner but are too big or to time consuming to let me build them locally on my tiny 1050TI gpu. I’m not upset with the limitation as I had my workflow in mind when I bought it but now that I’m stuck at home and not hauling the laptop everywhere I’d probably go the heavy duty route vs the lightweight route for the reasons expressed previously.
Ok one more question... Do you have a specific role name for a developer who specializes or who’s job entails mostly working with deep learning models to build products around them or are you using the same titles as you’d use for any task a developer may do? I personally think we’re not highlighting this developer role enough as a lot of jobs are more in this realm than the more research oriented realm.
Cheers,
Christopher
PS. Darn cute dog. :)
Hi jeff, i couldnt find the available options of linux in the preinstalled system. i was placing an order directly on the official website. if there is no way to customise preinstalled linux, is it the way that after linux installed by myself, i download the nvidia drivers as well as wifi drivers?
waiting for your dedicated video sir on what to do after installing ubuntu for Datascience distro!
i really hope your new videos could also be using windows.. i also expect to have some object detection video with yolo and tensorflow.
Most of my videos are Windows. Most of the actual code, though, can run in either.
But can it run Crysis? 🤔
Yes please, a video about having the various drivers and containers
Hah, yes RGB, maybe a guilty pleasure. And it does look cool on some of my lenses that can do the blurry background thing.
Lenovo P53s Laptop ~ i7-8665U ~ 16Gb ~ 512Gb SSD ~ nVidia P520. This laptop has been perfect for me and still is. It's perfect.
I have purchased 4 month before machine is good... but service is very poor and I did not get what they (sale person) promise me... never buy this product in India.
Maybe you should clarify that it's 12 logical cores and not 12 physical cores.
I was wondering if you could make a more detailed video on how to actually utilize the special machine learning hardware (like tensor cores)? From what I can tell, there aren't any videos that show how to make use of this new stuff. Is it already baked into Cuda and popular machine learning frameworks?
where to buy p53 with preinstalled fedora
Could u please do a review on T15G with Xeon?
Hi Jeff, I searched everywhere for lenovo laptop with preloaded Ubuntu, i couldn't find anywhere. Are they selling it?
Maybe this is something new. I will check the next time I discuss with them.
Thanks for the latest new.I looked into Thinkpad P series for buying one,but couldn't find one with Ubuntu OS.I think its not yet released.
They do not sell them with Ubuntu. You install the OS yourself. Ubuntu is an open source software so you can easily download it and install
@@MrGrowthAndChange What about all the tools and softwares he talked about,with all the dependencies correctly installed?
I have RTX2080 inside the box for 3 months ... Last time I configure tensorflow with GPU on ubuntu I had to format the harddrive 3 times . So Iam looking to be free to lose some days to install the board