Using a Laptop with an eGPU Enclosure: A Performance Comparison

It's no secret that using a laptop with an eGPU enclosure results in Thunderbolt and CPU bottlenecks, but just how much performance is being left on the table compared to putting the same graphics card in a desktop PC? In this article, we'll explore the performance differences between a laptop with an eGPU enclosure and a desktop PC, and how these bottlenecks affect gaming performance.

The author has compared their i9-10900K system against the Razer Blade Stealth 13, connected to the Mantiz Saturn Pro Gen II eGPU enclosure. The test machine was equipped with MSI's RTX 3090 GamingX Trio graphics card, which helps reduce GPU bottlenecks and reveal CPU and Thunderbolt limits. This laptop goes for around $1600 USD, which could definitely be used to build a nice gaming PC like the author's own system.

However, when you factor in the cost of the eGPU enclosure, which is around $300, the total cost comes close to $1900. This is still a significant investment, but one that may be worth it for those who value portability and want to play 4K games on their laptop. The author notes that while this setup is expensive, it's not without its benefits - with good settings, the laptop was able to deliver smooth performance in games.

The Performance Differences

In terms of gaming performance, the desktop PC was significantly faster than the laptop with eGPU enclosure. On average, the desktop system was a massive 127% faster on average in terms of average frame rate, with some games seeing as much as an 80% boost in Call of Duty. Control was still the best result, performing far better than any game on the laptop.

The performance differences were most noticeable at higher resolutions, where CPU and Thunderbolt bottlenecks started making less of a difference. At 4K, the desktop machine was still more than 50% faster than the laptop with the same graphics card. This is because the GPU was able to handle the increased load without hitting the limits of the eGPU enclosure.

Outside of gaming, the author found that the scores from the V-Ray benchmark were quite close between the two systems. The desktop PC was only around 2% better, which suggests that there may not be as much difference in more compute-heavy workloads. In the Blender OpenData benchmark, we're just seeing the desktop complete the task a few seconds faster, so the eGPU can be quite capable in heavy GPU rendering tasks that are likely to not smash the Thunderbolt link and processor super hard.

DaVinci Resolve was doing much better on the desktop, which the author attributes to the processor matters more in this workload. The overclocked 10900K is on a different plane of existence to the quad-core chip in the Blade Stealth. This highlights that while the eGPU enclosure can handle heavy GPU workloads, it's not without its limitations - and the CPU performance may still be a bottleneck in certain tasks.

The author concludes that the results are very interesting and not at all what they were expecting. The use of an eGPU enclosure is definitely a niche setup, but one that can deliver impressive performance for those who value portability and want to play 4K games on their laptop.

Overall, the use of an eGPU enclosure like this is still a relatively expensive proposition, but it's not without its benefits. For those who need a system that can handle both gaming and productivity tasks, the author suggests considering a mid-range gaming PC or a smaller laptop with good graphics card - there are plenty of options available at different price points to suit different budgets and requirements.

The Use of an eGPU Enclosure: A Niche Setup?

The use of an eGPU enclosure is definitely a niche setup that requires a fair bit of money to get started. However, for those who just want one system to do it all - value portability, have the money, and want to play 4K games on their laptop - this may be an option worth considering.

While the author notes that the CPU differences between these two setups are completely unfair, they suggest checking out another video if you want to see how the CPU in the Razer Blade Stealth compares with the Dell G7. The results are very interesting and not at all what was expected. In this comparison, the 6-core 10750H with higher power limit and higher clock speed is a much more powerful processor than the overclocked 10900K.

In any case, it's clear that the use of an eGPU enclosure can offer impressive performance for those who need a system that can handle both gaming and productivity tasks. Whether or not this setup is worth it will depend on individual needs and requirements - but one thing is certain: the results are very interesting and not at all what was expected.

"WEBVTTKind: captionsLanguage: enIt’s no secret that using a laptop withan eGPU enclosure results in Thunderbolt andCPU bottlenecks, but just how much performanceis being left on a table compared to puttingthe same graphics card in a desktop PC?I’ve compared my i9-10900K system againstthe Razer Blade Stealth 13, connected to theMantiz Saturn Pro Gen II eGPU enclosure.I’m testing both with MSI’s RTX 3090 GamingX Trio graphics card to reduce GPU bottlenecksand reveal CPU and Thunderbolt limits.This laptop goes for around $1600 USD, whichyou could definitely use to build a nice gamingPC like I’ve got here instead, plus you’vegot to factor in around $300 for the eGPUenclosure, so we’re almost at $2000 herebefore even buying a graphics card.There are of course cheaper Thunderbolt capablemachines though.The reason I’m using the Blade Stealth isthat due to it having Ice Lake with an integratedThunderbolt controller, it actually ends upperforming better in an eGPU setup than morepowerful Comet Lake H based laptops, likethe Dell G7 with 6 core 10750H processor.Both setups were tested with the latest versionof Windows and same Nvidia drivers, let’sget into the results, starting with 10 gamesfollowed by content creator workloads after.Red Dead Redemption 2 was tested with thegames benchmark tool.I’ve got the laptop plus 3090 eGPU shownby the red bars, and the desktop with same3090 shown by the purple bars.I’ve also tested three resolutions, whichare listed on the left with 1080p at the bottom,1440p in the middle, and 4K up top.This game was around 100% faster on the desktopat 1080p and 1440p, then a little lower witha 72% lead at 4K.Battlefield 5 was tested in campaign modeby running through the same mission on bothmachines multiple times.Even the 1% lows from the desktop are absolutelydestroying the average FPS from the laptopand eGPU, but hey at the same time, to befair to the laptop it is still managing toplay fine above 60 FPS even at 4K ultra settings,which you can’t really do on most laptopswithout an eGPU.Shadow of the Tomb Raider was tested withthe games built in benchmark.No question the desktop was far ahead here,with a 143% higher average frame rate at 1080p,though this lowers to a 45% lead at 4K, presumablyas the GPU is better loaded here so the limitationsin CPU and Thunderbolt start to become lessof a factor.Let’s start with RTX off results in Control.Again the desktop was smashing the laptopat lower resolutions, this game had the biggestdifference out of all 10 titles tested, withthe desktop more than three times faster thanthe laptop plus eGPU at 1080p.Once more at 4K the difference is far lesspronounced, just a 30% higher frame rate fromthe desktop now.With RTX on and DLSS enabled though, the desktopsees bigger gains than what we just saw.At 4K now it’s 86% ahead of the laptop andeGPU setup, so based on this it would seemthat ray tracing doesn’t work quite as wellin the eGPU compared to a full blown desktopPC.Metro Exodus was tested with the games benchmarktool.This was one of the smaller differences outof the 10 games tested, but still big improvementson offer from the desktop PC, where the 1%lows were near the averages from the laptopat 1080p and 1440p resolutions.Death Stranding was slightly above the 10game average at all resolutions, and thatmeans massive frame rate improvements withthe desktop, the 1% low on the desktop at1080p is getting close to double the averageframe rate from the laptop, crazy stuff.Assassin’s Creed Odyssey was tested withthe games built in benchmark.Another easy win for the desktop machine,so let’s keep moving on.Call of Duty Modern Warfare saw the smallestdifference between the two at 1080p and 1440presolutions in terms of a percentage.Just let that sink in, because looking atthe results here we can see there’s stillquite a big difference.Rainbow Six Siege was tested with Vulkan usingthe games benchmark tool.For some reason the eGPU setup was havingheavy stuttering issues at the start of thebenchmark, and that’s why it’s 1% lowsare way down.I replicated this on two different laptopswith different enclosures, so it seems tobe some eGPU specific issue, though it wasalso the only Vulkan title tested so thatcould be part of it.The eGPU setup was getting smashed in thewitcher 3 by the desktop, but again like manygames tested it’s important to keep in mindthat being able to game at 4K max settingswith the ultrabook in this setup is stillquite impressive.These are the performance differences in allgames at 1080p.On average the desktop system with superiorhardware was a massive 127% faster on averagein terms of average frame rate.Even worst case we’re looking at about an80% boost in Call of Duty, while Control wasmore than three times as fast.At 1440p the desktop PC is now just under100% faster on average when compared to thelaptop eGPU setup.Control was still the best result, performingfar better, but even worst case in these gameswe’re still seeing more than a 60% higherframe rate from the PC.I’ve kept the same X Axis as the other graphsfor 4K to give you an idea of how this compares.I believe we’re seeing the margins betweenthe two decrease at higher resolutions asCPU and Thunderbolt bottlenecks start makingless differences, a result of the GPU gettingloaded up.In any case at 4K the desktop machine wasstill more than 50% faster than the laptopwith the same graphics card.Outside of gaming, I found the scores fromthe V-Ray benchmark quite close.The Desktop PC was only around 2% better here,so it doesn’t seem that there’s anywherenear as much difference in more compute heavyworkloads.This sort of setup could be much more viableif you’re not just planning on gaming.Likewise in the Blender OpenData benchmark,we’re just seeing the desktop complete thetask a few seconds faster, so the eGPU canbe quite capable in heavy GPU rendering tasksthat are likely to not smash the Thunderboltlink and processor super hard.DaVinci Resolve was doing much better on thedesktop, I’m guessing in this video editingworkload the processor matters a bit more,and that overclocked 10900K is on a differentplane of existence to the quad core chip inthe Blade Stealth.Yeah, so as expected in games the frame ratesfrom the desktop system are much better whencompared to the laptop with eGPU, but at thesame time though I think it’s impressivewhat this was able to deliver.This is a lower powered 13” machine andwe were able to basically play 4K games withgood settings without much problem.It was definitely usable.In the end, I think it really comes down tohow much you value portability.Good luck taking something like this to theoffice or work compared to this laptop andthen just docking with the eGPU when you gethome.In many cases it’s cheaper to buy a midrange gaming PC and if you need portabilitythen maybe a smaller laptop, or maybe justa mid range laptop with good graphics card,either way there are a few different optionsso you can pick the one that suits your budgetand requirements.An eGPU setup is definitely niche and requiresa fair bit of money to get started.But for those that just want one system todo it all, value portability, and have themoney, an eGPU setup like this is definitelyan option.Now obviously the CPU differences betweenthese two setups is completely unfair, socheck out this other video if you want tosee how the CPU in the Razer Blade Stealthcompares with the Dell G7, as that’s usingthe 6 core 10750H with higher power limitand higher clock speed.The results are very interesting and not atall what I was expecting.\n"