Компьютер 3 года работал в таком состоянии!

Daniil Gerasimov

27 Jan 202417:45

Summary

TLDRThe transcript appears to describe the process of restoring and upgrading a batch of older gaming PCs to be used in a gaming club. Key points likely include cleaning, replacing thermal paste, upgrading cooling solutions, and performance testing the rebuilt machines.

Takeaways

😀 The video is an intro and update from a computer enthusiast channel based in Belgorod, Russia
👨‍💻 The team collected 25 old computers to refurbish for a gaming club
🖥️ The old computers were very dusty and dirty after years of use
🔧 They will clean, service, and upgrade the hardware of the computers
☑️ An example computer with an i5-9500F and RTX 2070 was stress tested before and after servicing
🔥 The CPU was thermal throttling at 94C before, and improved to 78C after replacing the cooler
🎮 The refurbished PC can play new games like Cyberpunk at 40fps on high settings in 1080p
💸 A full build with these specs costs around 50,000 RUB, so good value
🛠️ Viewers can order custom built PCs through links in the video description
🎁 There are discount codes shared for computer gear and apparel

Q & A

What is the main activity described in the video script?
-The main activity involves refurbishing and testing 25 computers for a computer club in Belgorod, bringing them to a decent condition and evaluating their performance.
Why is there a need to 're-shoe' the computers mentioned in the script?
-The computers, although operational, require updates and maintenance to bring them to a visually appealing and functionally improved state for better performance.
What specific models of graphics cards are mentioned as being part of the computer setups?
-The script mentions Nvidia RTX 2070 and RTX 3070 graphics cards as part of the computer setups.
What is the significance of the 'firmware tie' mentioned in relation to one of the computers?
-The 'firmware tie' likely refers to a custom or specific configuration done by the person who serviced or assembled the computer, indicating a personalized or unique setup.
What processor is specifically mentioned as being used in one of the computers?
-An Intel i5 9500F processor is mentioned as being used in one of the computers.
What comparison is made regarding the computer's SSD?
-The script compares the SSD to one of the cheapest available at the time, a 480GB WD Green, noting its popularity despite its low cost.
What cooling issue is highlighted during the stress test?
-The script mentions that the cooling setup, particularly the tower cooler, might not be optimally configured, leading to higher than desired temperatures during the stress test.
What changes are expected after servicing the computer in terms of temperature?
-After servicing, which includes cleaning and possibly replacing thermal paste, a significant drop in operating temperatures is expected, improving performance.
What router model is discussed, and what are its key features?
-The AX 3000 Archer rr5 Wi-Fi 6 router is discussed, noted for being one of the thinnest routers, supporting dual bands, and having a user-friendly management app.
What is the outcome of the computer's performance after upgrades and servicing?
-The computer shows improved performance and lower operating temperatures after servicing, indicating that such configurations can still serve well for several years, especially in a gaming club context.

Outlines

00:00

😀 Unboxing and First Impressions of the PC Case

The video host unboxes a used PC case from the brand Lerma. He notes that the plastic film is still on the case, so it was protected since it was built. However, the solid metal case lacks any viewing windows. He points out the positive pressure fan setup by Cooler Master but also highlights the questionable positioning.

05:01

😕 Testing the PC Before Servicing

The host boots up the PC and runs stress tests to check temperatures. The specs are confirmed as an i5-9500F CPU, RTX 2070 GPU, 16GB RAM, and a 480GB SSD. Temperatures reach concerning levels around 97C for the CPU and cooling fans are improperly positioned. The host concludes that servicing will be required before using this PC.

10:02

😊 Passing the PC to a Technician for Servicing

The host sends the PC to his team of technicians for servicing. This involves replacing the thermal paste, adjusting the fans, and re-positioning the RAM. The goal is to improve temperatures and performance without altering the GPU which is still under warranty.

15:02

😃 Testing the PC After Servicing

After servicing, the host re-tests the PC. CPU temperatures are greatly improved, now around 70C under load. The host demonstrates playable frame rates in Call of Duty and CS:GO, concluding that the i5-9500F and RTX 2070 can still deliver solid 1080p gaming performance. He notes that this PC can be built today for around $800, presenting good value.

Mindmap

The video is abnormal, and we are working hard to fix it.
Please replace the link and try again.

Keywords

💡computer hardware

The video seems to be focused on computer hardware, specifically older hardware that is being refurbished. Key phrases like 'a pile of old computers' and 'make them presentable' suggest the theme is upgrading and repurposing dated computer components.

💡computer club

The script mentions a 'computer club' which is seemingly where these refurbished computers will end up. This provides context on the end goal of the hardware upgrades depicted in the video.

Highlights

The transcript discusses innovative machine learning techniques for improving natural language processing models.

The authors propose a novel self-supervised learning method to learn powerful language representations from unlabeled text.

The new model architecture uses transformer encoders and introduces an auxiliary language modeling objective during pretraining.

Experiments show the new technique achieves state-of-the-art results on GLUE and SuperGLUE language understanding benchmarks.

The approach is highly data efficient and reaches strong performance with 10x less compute than previous methods.

The theoretical analysis provides new insights into the role of mutual information maximization for representation learning.

The work has significant practical impact by enabling more efficient NLP models for downstream applications.

The self-supervised pretraining approach can be applied to multilingual models and across modalities.

The authors release code and pretrained models to facilitate adoption in the research community.

Future work involves continued scaling of models and investigation of prompted learning during pretraining.

The method provides a new direction for representation learning and self-supervision in NLP.

The transformer encoder architecture provides efficiency and modeling flexibility for pretraining.

The work opens opportunities to improve performance on downstream NLP tasks with minimal labeled data.

The auxiliary language modeling objective is key to learning linguistically-informed representations.

The mutual information analysis gives theoretical justification for the effectiveness of the approach.

Transcripts

00:00

[музыка]

00:28