I need assistance understanding the cause of your system's slowness.
I need assistance understanding the cause of your system's slowness.
Hey everyone,
I've noticed my systems have been running sluggishly for about a year now. I'm referring to systemS since I initially had an Alienware Area 51 R5 in 2018. Over the last year, I've replaced nearly all components except the PSU—though I'm not sure if that's the real cause of the lag, but it could be.
The upgrade came after a suggestion from my programmer friend, who thought it might be related to the PCIE lanes due to multiple NVMe drives in my setup. I upgraded the GPU first, then moved on to the motherboard and CPU, followed by RAM. Recently, I swapped out two of my 2TB NVMe drives for a single 4TB drive, bringing my total NVMe drives down to just two 4TB units.
Current hardware:
- MSI Z790-A WiFi Motherboard (4x M.2 slots)
- [Link to specs](https://www.msi.com/Motherboard/PRO-Z790...cification)
- Zotac RTX 4070ti GPU
- Intel i5 13600k CPU
- 64GB Corsair Vengeance Pro DDR4 RAM (4 x 16GB dimms)
- 2x Crucial P3 Plus 4TB PCIe Gen4 NVMe M.2 SSDs
- 2TB Segate HDD
- 8TB Seagate HDD
- 2TB Crucial SSD
I just completed a fresh Windows 10 Pro 22H2 installation yesterday—this is the third time I've experienced these slowdowns.
When I mention slowdowns, it's mainly during tasks like running Unreal Engine 5, 3D Studio Max, or streaming video. Everything was smooth before; now, in Unreal I experience frequent frame drops, dropping from 120 fps to 22 fps for a few seconds, no matter the scene. In 3ds Max, I see constant stuttering while scrolling in the UI.
Streaming is a bit better, but I sometimes encounter lag spikes where videos freeze, and with Netflix, Amazon, or Disney Plus, my monitors freeze briefly before returning to normal. With Netflix, it stutters constantly when scrolling videos, unlike the other two services. I haven’t noticed any issues in games, though I haven’t played much recent AAA titles, so that’s probably the reason.
I’m guessing the problem is still related to the PCIe lanes or the GPU I purchased. To be honest, the first few months after buying the GPU were great (before any other upgrades), and when I play Dead By Daylight everything runs smoothly.
CPUZ screenshot: [link](https://imgur.com/M0knIFR)
View: [image link](https://i.imgur.com/M0knIFR.png)
The build details were not specified regarding the PSU make and model. A link to the RAM kit would be appreciated. The PCB revisions for the RAM modules would also be useful.
I recently installed Windows 10 Pro 22H2 three times this week due to performance issues. Could you provide a link to the RAM kit? An image of the PCB changes for the RAM sticks would greatly assist us.
The installation was completed yesterday, and I've noticed the slowdowns have returned after the third attempt. Where did you obtain the OS installer? Was it installed offline?
I suspect the problem might still be related to the PCIE lanes or the GPU I purchased. Although the initial months post-GPU upgrade were smooth, especially when playing Dead By Daylight, the recent instability is concerning. The manual doesn't mention any lane sharing issues.
It's just the PSU that came with my Alienware, so I assume it's a Dell. I don't know the exact model. I might open it up tomorrow to check, but it performed well for five years in my Alienware. I believe it's an 800 Watt unit, though I'm not sure. I could look again if you think it's important.
PCIe connections from the CPU and chipset are completely separate, so they wouldn't cause bottlenecks. There are no GPUs with more than 16 PCIe lanes and no NVME SSDs with more than 4. The CPU actually provides 24 PCIe lanes, but only 4 are used for I/O and chipset functions.
I believe the absence of discussion around PCIe lanes in theory is a distraction.
Occasionally, adding another M.2 drive can disable a PCIe slot, or an x16 port might revert to x8 operation, though this shouldn’t impact performance.
I’m beginning to suspect the issue might be linked to the RTX 4070—your graphics may stutter or disappear. Do you still have the previous GPU? If yes, remove the 4070 and install the older graphics card (or any compatible one) to test stability. It’s simple to adjust RAM or switch the CPU if needed. Gradually revert the system to its original parts to determine if the problem persists. If it does, it might relate to your software, operating system, or boot SSD.
It may not apply to Alienware models, but many Dell office machines include proprietary power supplies that differ from the standard ATX PSU used by most home PC builders.
If your Dell PSU lacks a 24-way power connector for the motherboard and also doesn’t provide a 4 or 8-way ATX12V power connector (for CPU power), you’re dealing with a non-standard proprietary unit. This can complicate PSU replacement.
Lower-end PSUs typically come with 1 to 3-year warranties, mid-range with 5 years, and high-end with 10 or 12 years. Check your Alienware’s warranty details—it’s likely not valid beyond five years.
Even if the PSU still functions for another five years, it could fail sooner. High-end models with 10-year warranties use superior components, such as 105°C capacitors instead of the standard 85°C ones.
As per the Tech Reviewer article referenced, a mid-range system like yours should ideally have a 750W power supply (their stated maximum is 588W). Adding about 20W for hard drives should keep an 800W unit sufficient, especially if it’s a quality gold model.
In most cases, your computer will consume much less power—around 100W when idle.
Power Supply Recommendations for RTX 4070 Ti and Mid-Tier Parts
Mid-Tier Components
Peak Power Consumption
RTX 4070 Ti GPU
285 watts
Mid-Tier CPU (e.g., Intel Core i5 13600K)
181 watts
Motherboard
80 watts
Two M.2 or two 2.5" SSDs
18 watts
Two Case Fans (120 mm)
12 watts
Two CPU Fans (120 mm)
12 watts
Overall Estimate: 588 watts
Recommended Power Supply Wattage:
750 watts
https://www.techreviewer.com/best-tech/n...er-supply/
Dell and Alienware generally rely on Greatwall as a reliable PSU maker. When retail motherboards are installed, the PSU meets ATX specifications. However, Dell usually maintains minimal requirements for many configurations, which could put additional strain on the PSU.
Your RAM is not performing as expected. It operates at the lowest default performance level. You might need to increase the voltage to achieve better speed or stability. Consider running memtest86+. It launches from a USB drive and does not require Windows. You can obtain it from: Memtest86+ | The Open-Source Memory Testing Tool. If you complete all tests without errors, your RAM appears to be functioning correctly. Additional passes may reveal problems but will take longer. This should only be done if you strongly suspect a RAM issue.
How can I have mismatched RAM when I bought two kits of the same? I believe I might be confusing things and assuming I'm still using the Alienware RAM. The only part remaining from the Alienware is my PSU right now. I'll run the diagnostic, thank you for the link.
I started by opening the case, but there’s nothing visible on the PSU to identify its wattage or manufacturer. It looks like the label is either hidden under the case or on the side covered by the separator. I’m not trying to pull it out right now. If necessary, I’ll consider other solutions first. I don’t remember exactly, but when I saw the PSU and the Mobo, I was surprised they weren’t cheap, especially given Dell’s acquisition of Alienware.
If the system is underpowered, would that cause frame drops like this?
I’ve been leaning toward thinking the problem lies with the GPU since it’s only about a year old now. Unfortunately, I don’t have the original unit because I installed it in a friend’s system. I do have an older GPU—a GT-710—that could be used if needed.
Beyond running tools like Furmark or benchmarking software such as Unigine, do you know of any methods to assess GPU stability? The thing that confuses me is I haven’t noticed any frame rate issues when playing Dead By Daylight, which makes me wonder if the GPU isn’t the issue. Alternatively, it could be that DBD isn’t demanding enough to cause problems.
I might be mistaken about the timing of these components now. It seems my GPU is less than a year old, and I’ve been dealing with this for some time. If it’s a software problem, I thought reinstalling Windows would help. Hmm...
I checked my receipts and noticed I upgraded the case, mobo, and CPU on November 7th, 2023, while my GPU was installed on May 12th, 2023. It seems the problems began about 3 to 5 months after the GPU installation, which would place them around August to October of last year—probably why I thought the GPU wasn't yet a year old.