The Alienware m15 R5 Update: A Mixed Bag for Performance

Just a 1 to 3 FPS difference may seem like a minor change, but it's worth noting that this discrepancy can add up, especially when it comes to demanding games. In Assassin’s Creed Valhalla, the updated system was able to maintain its performance lead at lower settings presets, while high settings saw only a small gain in favor of the updated system. However, when ray tracing was enabled, the update started to show its benefits, with an almost 5% boost to average FPS.

The difference is not limited to just Assassin’s Creed Valhalla, as testing also revealed that Control was able to maintain its lead at high settings, albeit with only a 3 FPS advantage. Additionally, Battlefield V showed some unusual results, where the updated system actually performed worse than before. This anomaly was seen across all setting levels, making it unclear why this specific game saw a decrease in performance.

On the other hand, Microsoft Flight Simulator saw an improvement in both average FPS and 1% lows, with the update providing a slight edge in performance. However, results for Watch Dogs Legion were less convincing, as both versions of the game performed similarly at high settings. The Witcher 3 also showed no significant changes, with only minor variations seen across different setting presets.

One interesting observation was made during testing, where CS:GO saw some improvements after the update, particularly in terms of 1% lows. However, the gain was limited to around 2%, making it unclear whether this was enough to justify the update.

The results from these tests are a mixed bag for the Alienware m15 R5's performance, with some games seeing significant gains and others performing worse or similar. This raises questions about why some games saw an improvement while others did not.

Bypassing the iGPU with advanced Optimus settings should provide better performance in certain titles, as this technology allows the system to utilize the full capabilities of the GPU. However, it's unclear whether this would be enough to offset the performance differences seen between the updated and non-updated systems.

The battery life also saw no significant changes after the update, with The Witcher 3 lasting around a couple of minutes less than before. This is within margin of error and not enough to draw any solid conclusions from.

Overall, it seems that while there are some minor gains to be had from the update, they do not amount to anything major. The updated system still struggles to match the performance of a dedicated gaming laptop with an RTX 3070 GPU. However, Alienware customers should now be getting full utilization of their hardware, as the problem has been resolved.

As I continue working on my review of the Alienware m15 R5 gaming laptop, I encourage readers to subscribe to my channel for updates and behind-the-scenes content. You can also join me in Discord and support the channel by becoming a patron on Patreon. In the meantime, be sure to check out some of my other videos over here.

"WEBVTTKind: captionsLanguage: enThe Alienware m15 R5 gaming laptop was missingabout 10% of its CUDA cores when it initiallylaunched, I’ve covered this in another videolinked down in the description, but now Alienwarehave released an update that unlocks the restof those CUDA cores, however it’s not exactlydoing quite as well as I expected.I’ve retested the m15 R5 in 10 differentgames to show you how well it performs bothbefore and after the update, and I’ve alsotested out thermals so we can get a completepicture of how this laptop is supposed toperform.These are the specs of my m15 R5, I’ve gota Ryzen 9 5900HX processor, Nvidia RTX 3070graphics, 16 gigs of memory and a 1080p 165Hzscreen.It’s worth noting that the 1080p 165Hz screenI’ve got means that we’re stuck with optimus.The 1440p 240Hz or 1080p 360Hz options comewith advanced optimus and G-Sync, so I wouldexpect those to perform even better to whatI’m showing here.Here’s how things looked before the update,so both system information in Nvidia’s controlpanel and GPU-Z show 4608 CUDA cores, andthen once the new VBIOS update was installedwe can see the correct number of 5120 coresare shown.So I basically downloaded more CUDA cores,hopefully not too long before we can downloadmore RAM.The update wasn’t available for me throughthe Alienware update software, you have tomanually go to the website and get it.Hopefully this changes, otherwise I’m surea lot of machines will probably never getpatched.Here’s the download page for the updateon Dell’s website.I found it interesting that under fixes & enhancementsit notes “corrected the CUDA core informationin system information from 4608 to 5120 inthe Nvidia control panel application”.Based on that description it kind of justmakes it sound like this is a superficialfix rather than actually modifying the amountof available CUDA cores, so let’s checkout some actual performance tests and seewhat the differences were before and afterthe update.These are the screenshots from hardware infowhen running the Heaven GPU benchmark in thesame room with a 21 degree Celsius ambientroom temperature.The TDP and clock speeds are basically thesame and within margin of error both beforeand after the update, but check out the temperaturedifferences.After the update was running 6 degrees cooleron average over the course of an hour.The GPU hot spot temperature was also around7 degrees cooler after the update, so sameworkload in the same conditions, literallythe only change was the VBIOS update.So what’s going on here?I don’t know for sure, but this is whatI think is happening.On average the RTX 3070 was using the same122 watts both before and after the update,but before the update that same amount ofpower was available for fewer CUDA cores.After the update that same amount of powerthen becomes available for 10% more CUDA cores,and CUDA cores take up physical space on thedie, so in theory more CUDA cores equals morephysical surface area, so my thought is ifyou’re spreading that same amount of powerout over a bigger area then it could leadto lower GPU hot spots.Again I can’t say for sure, this is justmy best guess based on the information available.Now although things are clearly differentwith the GPU only workload, there’s basicallyno change when the processor is also underload, like when we’ve got an actual gamerunning for instance.These are the temperatures for both the CPUand GPU when under combined stress test, soboth processor and graphics are fully loadedup, both before and after the update we’relooking at basically the same temperatures.Likewise the clockspeeds are essentially thesame too, technically after the update wasaround 20MHz slower on both the CPU and GPU,but this is quite small and within marginof error.Again the power levels for both CPU and GPUwere about the same in these CPU plus GPUstress tests, no obvious important differences.So just to summarize things so far, afterthat VBIOS update in a GPU only stress testI was seeing lower temperatures, but thatwas about the only difference, and then whenthe laptop was under combined CPU plus GPUstress test there was basically no changesto note.Just before the games, let’s take a lookat 3DMark results, as these are typicallyGPU heavy.I’ve got the updated results shown in purple,and in all tests the updated system is scoringbetter than before the update, though thedifferences aren’t that big.The timespy graphics score is about 3.5% higherafter the update, while the Firestrike graphicsscore has a 1.8% improvement.The Port Royal ray tracing test on the otherhand was around 6.7% higher after the update.So after the update there’s clearly a boostin these GPU only workloads, but what aboutactual games?The results surprised me and weren’t quitewhat I was expecting, so let’s compare 10games both before and after the update.Cyberpunk 2077 was tested in the exact samemanner both before the update in the red bars,and after the update in the purple bars.Regardless of the setting preset, or whetheror not we’re using ray tracing, the updatedlaptop is performing better.The new vBIOS update was reaching almost 13%higher average FPS at low settings, whileultra settings was 4.5% ahead.It’s not all gains though, Red Dead Redemption2 was tested with the games benchmark, andthe system before the update was actuallyslightly ahead between low and medium settings,granted it’s only a small amount and forthe most part a couple of FPS is within marginof error.The updated laptop was ahead at ultra settings,and generally I’d expect higher settingpresets that are more GPU heavy to see thebenefits of the CUDA core change.Things really start getting weird in Shadowof the Tomb Raider.Regardless of the settings in use the updatedsystem was consistently behind.Sure it’s often just a 1 to 3 FPS difference,but the laptop prior to the update was comingout ahead consistently.Assassin’s Creed Valhalla was similar, atmost setting presets it’s within marginof error, though for some reason high settingswas a fair bit different in favor of beforethe update.Control is a GPU heavy game, and at mediumand high settings the updated system is backin front, though it’s just a 3 FPS or sodifference, nothing major, putting high settingsaround 4% ahead after the update.Here’s how things look with ray tracingenabled, another workload that’s heavy onthe GPU, and this time the update was almostgiving us a 5% boost to average FPS.Performance improves with DLSS enabled, andat max settings the update is back to givingus around a 4% performance boost.Battlefield V was another weird one wherethe laptop was doing better before the updateat all setting levels.I double checked the results after the updateand got the same again, it’s only a smallchange, but again I’m not really sure whywe’d see lower performance with more CUDAcores available.Microsoft Flight Simulator was always slightlyahead with the update both in terms of averageFPS and 1% lows, regardless of setting presetin use, but it’s often quite a small difference.Watch Dogs Legion doesn’t really tell usanything useful, both were about the sameat the highest ultra setting preset, whileresults at lower setting presets were goingone way or the other.Likewise The Witcher 3 was very similar beforeand after the update, no real changes andit varied slightly one way or the other basedon the settings in use.CS:GO was generally doing better after theupdate, the 1% low results were ahead withthe update at all setting levels, while averageFPS had the biggest improvement at max settings,though it’s less than a 2% gain.These are the differences out of all 10 gamestested at the highest setting level, thoughI’ve also included ray tracing and DLSSresults which are adding more results in favorof the updated system despite them comingfrom the same games.In any case out of this selection the updatedm15 R5 was best case offering 5% higher averageFPS in Control with ray tracing, while somegames were actually doing worse.Now it’s a small difference either way,and in general the trend is positive, butstill I wasn’t expecting to see some doinga little worse, granted it could be some otherissue with the testing, though I feel confidenton the testing both before and after updating.Testing at 1440p may have shown a bigger differenceas this would put more of a workload on theGPU, but unfortunately I’ve got the 1080pscreen model here.I guess I could have used an external screenin order to run at higher resolutions, butbasically I was just using the data that wasalready collected prior to the update.Those 10 games were just tested out for myusual game benchmark video, but that was beforeI found out about this issue, so instead ofjust scrapping that work and doing it allagain anyway I figured it would be usefulto show you the before and after differences.Given we actually saw worse performance inShadow of the Tomb Raider and BattlefieldV, compared to others the m15 R5 still isn’tstacking up well for an RTX 3070 gaming laptop,at least in these titles, but I should notethat the other screen options with advancedoptimus should perform better as they wouldbypass the iGPU.Battery life saw basically no change either,after the update running The Witcher 3 lasteda couple of minutes less, but this is marginof error stuff and not enough to draw anysolid conclusions from.So at the end of the day after the updatethe m15 R5 doesn’t get that much of a performanceboost.In general there is some improvement but nothingmajor.Personally I did expect a bigger differencethan this given 10% of the CUDA cores weremissing, but perhaps at least at 1080p itwasn’t enough to matter.In any case I think it’s good that Alienwarecustomers are now getting that full 3070 thatthey’re paying for, and that the problemhas been resolved.I’m still working on the full review ofthe Alienware m15 R5 gaming laptop, so ifyou’re new to the channel then get subscribedso that you don’t miss it.Come and join me in Discord and get behindthe scenes videos by supporting the channelon Patreon, and while you wait for the reviewcome and check out some of my other videosover here, I’ll see you over in one of thosenext.\n"