Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Fallout 4 can become heavily CPU-limited in dense scenes. The most useful examples are Corvega and Diamond City, where the game’s DirectX 11 rendering workload can overwhelm the CPU-side render thread before a modern GPU is fully occupied. The historical AnandTech Forums benchmark known as “[Part 3] Measuring CPU Draw Call Performance in Fallout 4” is valuable evidence of that behavior—but it is not a controlled modern CPU ranking.
The original results favored strong single-thread performance, fast memory, and, in the tested configurations, Nvidia’s DX11 driver path. Later community submissions show large gains from newer CPUs, memory platforms, and GPUs, but they cannot be combined into a reliable generational chart because the game versions, drivers, graphics cards, RAM settings, overclocks, mods, and measurement methods changed.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Fallout 4 (PC) | $19.95 | Buy on Amazon |
| 2 |
|
Fallout 4 Game of The Year Edition - PC | $49.99 | Buy on Amazon |
| 3 |
|
Fallout 4 | Anniversary Edition | XBOX and Windows Digital | $59.99 | Buy on Amazon |
| 4 |
|
Fallout 3 - PC Game of the Year Edition | $9.55 | Buy on Amazon |
| 5 |
|
Fallout 76: Wastelanders - PC | $33.99 | Buy on Amazon |
What the benchmark is measuring
A draw call is a command sent by the CPU-side rendering code to the graphics API instructing it to render a mesh, material, object, or batch. Draw calls are not the same thing as polygons, NPCs, objects, or visible pixels. A single object may require several calls because of separate meshes, materials, lighting, shadows, transparency, or effects.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Draw calls become a performance problem when the CPU’s rendering thread and graphics driver cannot submit work quickly enough to keep the GPU busy. The result is a CPU or render-thread bottleneck: lowering resolution may produce little improvement, while a faster CPU or lower-overhead driver path can raise frame rate.
#1 Best Overall
Draw-call count is only one part of the workload. Engine scheduling, scene complexity, driver behavior, memory latency, synchronization, streaming, and GPU work also affect the result. A higher reported count does not automatically mean a slower system, and a frame-rate difference does not prove that draw-call submission alone caused it.
Why Corvega and Diamond City matter
The original benchmark, started on June 13, 2018, selected two repeatable but demanding locations:
| Location | Approximate reported load | What it stresses |
|---|---|---|
| Corvega | About 11,000 draw calls | A particularly dense rendering scene |
| Diamond City | About 8,000 draw calls | Heavy rendering combined with NPC activity |
| Player settlements | Up to about 20,000 in some cases | Extreme, but difficult to reproduce because every settlement is different |
These are stress tests, not averages for the whole game. Diamond City’s count can vary with NPC positions and other simulation state; repeated testing reported roughly 7,900 to 8,400 calls. A settlement benchmark is only meaningful if the exact save, construction layout, player position, and game state are distributed with the test.
Recommended Free Tools
What the original Part 3 results reported
The original comparison reported that early Ryzen systems were approximately 21–23% slower than Skylake/xLake systems in the selected draw-call-heavy Fallout 4 scenes. The author also argued that fast DDR4 and an Nvidia GPU could narrow the gap. This was a workload-specific observation, not evidence that Ryzen processors are generally poor at rendering or gaming.
Rank #2
- Freedom and Liberty! Do whatever you want in a massive open world with hundreds of locations, characters, and quests. Join multiple factions vying for power or go it alone, the choices are all yours
- Be whoever you want with the S.P.E.C.I.A.L. character system. From a Power Armored soldier to the charismatic smooth talker, you can choose from hundreds of Perks and develop your own playstyle
- New next generation graphics and lighting engine brings to life the world of Fallout. From the blasted forests of the Commonwealth to the ruins of Boston, every location is packed with dynamic detail
- Intense first or third person combat can also be slowed down with the new dynamic Vault Tec Assisted Targeting System (V.A.T.S) that lets you choose your attacks and enjoy cinematic carnage
The same tested configurations reportedly showed Nvidia cards delivering about 30% higher frame rates than AMD cards. The proposed explanation was lower driver overhead in this Fallout 4 DirectX 11 workload. That finding should not be generalized to all games, GPU generations, or APIs. DirectX 12 and Vulkan expose different driver and submission paths, and a result in one older engine does not establish a universal vendor hierarchy.
Forum participants also questioned whether the test isolated draw calls cleanly. Fallout 4’s engine behavior, incomplete multithreading, driver work, memory latency, and other API-specific effects may all contribute. The benchmark therefore measures a practical render-thread stress case rather than a laboratory-pure draw-call score.
Why memory tuning affected early Ryzen results
Early Ryzen platforms could be especially sensitive to memory latency, bandwidth, and Infinity Fabric settings. The thread includes testing in which performance rose from approximately 62 FPS with DDR4-3200 CL16 to 68 FPS with DDR4-3600 and optimized subtimings. That result came from a participant’s configuration and should not be treated as a guaranteed gain for every system.
The practical lesson is that memory can matter when the CPU render thread is the limit. Frequency, primary timings, secondary and tertiary subtimings, command rate, and—on applicable Ryzen platforms—fabric frequency should be recorded. Memory tuning does not isolate draw-call performance by itself, however. It changes the CPU’s overall ability to process the workload.
Rank #3
- ANNIVERSARY EDITION — Celebrate ten years of Fallout 4 with the complete post‑nuclear RPG experience, winner of over 200 Best Of awards including DICE and BAFTA Game of the Year.
- COMPLETE BASE GAME INCLUDED — Experience the full Fallout 4 main campaign as the Sole Survivor of Vault 111, shaping the fate of the Commonwealth through meaningful choices, factions, and exploration.
- ALL 6 OFFICIAL EXPANSIONS — Includes Automatron, Wasteland Workshop, Far Harbor, Contraptions Workshop, Vault‑Tec Workshop, and Nuka‑World for massive story, settlement, and gameplay additions.
- 150+ CREATION CLUB ITEMS — Expand your adventure with curated Creation Club content featuring unique weapons, power armor sets, building packs, paint jobs, player homes, and Dogmeat cosmetic variants.
- OPEN‑WORLD FREEDOM — Explore hundreds of locations, craft weapons and armor, build and manage settlements, and customize your playstyle using the S.P.E.C.I.A.L. system and perk progression.
Validate stability before benchmarking. An unstable memory overclock can cause crashes, inconsistent frame times, or silent data corruption while appearing faster in a short run. A stable slower configuration is more useful than an invalid high score.
Later community results show progress, not a clean ranking
The thread continued receiving submissions through September 2024. Several illustrate how much performance improved, but they are not directly comparable because the hardware and software conditions changed together.
| Reported system | Corvega | Diamond City | Important qualification |
|---|---|---|---|
| Ryzen 9 3900X, Radeon VII | About 57.8 FPS | About 66 FPS | Approximately 11,690 and 8,010 reported draw calls |
| Core i7-12700KF, Radeon RX 6800 XT | 87.8 FPS | 94.1 FPS | Compared by the poster with a Ryzen 7 5800X at 69.8 and 81 FPS |
| Core i5-12400, approximately 5.3 GHz, DDR5-6288, RTX 4090 | 134.5 FPS | 106.7 FPS | Aggressive overclock and memory tuning; not a stock comparison |
| Ryzen 7 7800X3D, DDR5-6000, RTX 4080 | 140.7 FPS | 114.2 FPS | 2024 community submission |
Different GPUs, drivers, operating systems, game configurations, clocks, memory timings, draw-call counts, and background applications are enough to invalidate a simple “fastest CPU” chart. The 7800X3D submission cannot be used to claim that it is a fixed percentage faster than the 5800X, and the 12700KF result cannot be separated from its Radeon card and platform settings.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteHow to reproduce the test responsibly
The original author later pointed readers toward a separate Fallout 4 draw-call benchmark thread containing configuration files, instructions, and saves. Because the complete walkthrough and required files are not reproduced here, the following is a controlled methodology rather than a claim of byte-for-byte replication.
Rank #4
- Fallout 3: Game of The Year Edition
- Freeze the software environment. Record the Fallout 4 edition, executable and patch version, Windows build, graphics driver, ENB version if used, mods, and load order.
- Use fixed saves. Load identical Corvega and Diamond City saves and use the same player position, view direction, and camera movement. Do not substitute a settlement unless the exact settlement save is shared.
- Keep the graphics configuration identical. Record resolution, refresh rate, V-sync, frame cap, preset, individual options, and all relevant INI files. A wrong or ignored INI can silently change the test.
- Control one hardware variable at a time. Use the same GPU when comparing CPUs, the same CPU when comparing GPUs or drivers, and document CPU clocks, SMT/HT state, power limits, GPU clocks, and memory settings.
- Confirm that the GPU is not the limit. At the selected settings, monitor GPU utilization and frame time. A GPU-limited run hides CPU differences.
- Capture draw calls and performance together. Record the capture tool and method. A screenshot of one FPS value is not enough to establish a result.
- Run multiple passes. Let the scene finish loading, then repeat the same route or observation several times. Discard runs affected by streaming or unusual NPC behavior, and report how many passes were used.
- Report frame-time data. Include average FPS and frame-time percentiles or 1% lows derived from a consistent capture interval. Average FPS alone can conceal stutter.
- Check stability. Test RAM and CPU settings before collecting data, and close monitoring or tuning utilities that may affect scheduling. One participant reported lower performance with Ryzen Master open, but that observation is anecdotal.
A useful result record includes:
| Field | What to record |
|---|---|
| CPU | Model, architecture, clock, core count, SMT/HT state, power limits |
| GPU | Model, driver version, clock behavior |
| Memory | Capacity, frequency, primary timings, subtimings, command rate, fabric setting where applicable |
| Software | Windows build, game build, ENB, mods, load order, INI files |
| Scene | Corvega or Diamond City, save identity, position, route, captured draw calls |
| Performance | Average FPS, 1% low or frame-time percentile, pass count, anomalies |
Diagnosing your own Fallout 4 bottleneck
Start with GPU utilization and frame-time graphs rather than assuming that every settlement or city slowdown is a draw-call problem.
- Likely render-thread limited: GPU utilization falls below full load, one CPU thread is heavily occupied, frame time rises in dense scenes, and lowering resolution produces little improvement.
- Likely GPU limited: GPU utilization remains near its practical ceiling, lowering resolution or expensive visual settings improves frame rate substantially, and CPU headroom remains.
- Likely mod or simulation limited: scripts, NPC behavior, physics, settlement logic, or a specific mod changes performance even when the scene’s rendering workload appears similar.
- Likely synchronization limited: V-sync, a frame cap, refresh-rate behavior, or an engine limit prevents the displayed FPS from revealing available CPU or GPU headroom.
Per-core utilization is more informative than total CPU utilization. A game can show moderate overall CPU usage while one important rendering thread is saturated. Conversely, a low GPU percentage does not prove that draw calls are the cause; streaming, scripts, synchronization, and engine stalls can also leave the GPU waiting.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the benchmark can—and cannot—prove
It can help reveal
- Relative performance in dense, CPU-limited Fallout 4 scenes.
- Sensitivity to CPU single-thread and render-thread throughput.
- The effect of memory latency and bandwidth on some platforms.
- Driver-path differences in a particular DirectX 11 game configuration.
- Whether a faster GPU matters after the CPU rendering thread is saturated.
It cannot reliably measure
- Average performance across the entire Fallout 4 world.
- Every mod list or settlement configuration.
- Loading times, script throughput, physics, or VR performance.
- Modern DirectX 12 or Vulkan draw-call behavior.
- General CPU or GPU performance across other games and engines.
Reducing visual quality may not solve a draw-call bottleneck if the game still submits essentially the same collection of objects. Conversely, a GPU upgrade can help greatly if the original test was actually GPU-limited. The correct upgrade depends on the measured bottleneck.
Free tools Windows power users keep installed
One-click scans. No signup required.
Common reasons results do not match
- Different game builds: patches can alter performance and invalidate direct comparisons.
- Different saves: NPC positions, loaded objects, and simulation state can change the scene.
- Different ENB or mods: post-processing and asset changes can alter both rendering and memory behavior.
- Mixed GPU vendors: driver overhead contaminates CPU comparisons.
- Different memory settings: frequency and timings can materially affect latency-sensitive results.
- Single-run testing: streaming and NPC activity can create false conclusions.
- Unstable overclocks: apparently higher performance may be invalid.
- Misread draw-call counts: a changed count may indicate a different scene or capture rather than a faster or slower CPU.
Practical recommendations
If an older PC struggles in Corvega or Diamond City, prioritize CPU render-thread performance and a stable memory configuration before buying a faster graphics card. On early Ryzen platforms, sensible memory tuning may help, but stability is more important than a headline frequency.
Best Value
- Bethesda Game Studios, the award-winning creators of Skyrim and Fallout 4, welcome you to Fallout 76, the online prequel where every surviving human is a real person. Work together, or not, to survive. Under the threat of nuclear annihilation, you’ll experience the largest, most dynamic world ever created in the legendary Fallout universe
- Reclamation Day, 2102. Twenty-five years after the bombs fall, you and your fellow Vault Dwellers—chosen from the nation’s best and brightest – emerge into post-nuclear America. Play solo or join together as you explore, quest, build, and triumph against the wasteland’s greatest threats
- Contains Digital Download Code. No Disc Included
- Must be redeemed on the Bethesda client
If GPU utilization is already high, a GPU upgrade or lower graphics settings may be appropriate. If the GPU is underused while one CPU thread is saturated, more GPU power is unlikely to solve the specific problem.
For modded settlements, reduce object density, simplify the settlement, isolate heavy mods, and test with a controlled save. Do not assume that a new GPU will fix a settlement whose primary limit is CPU submission or simulation.
For benchmarking, keep the game build, save, graphics card, driver, memory configuration, and measurement method fixed. Separate stock, overclocked, and memory-tuned results, and publish enough information for another reader to understand what changed.
Bottom line
Part 3 is useful because it demonstrates a real and repeatable Fallout 4 performance pattern: dense scenes can saturate the CPU-side rendering path, and the result is sensitive to architecture, memory behavior, and the DirectX 11 driver path. Its reported Intel, Nvidia, and early-Ryzen differences are historical findings from specific configurations—not permanent laws about CPUs, GPUs, or vendors.
Use Corvega and Diamond City as controlled stress tests, not as a universal leaderboard. Repeated runs, fixed saves, complete configuration records, GPU and per-thread monitoring, and frame-time data are what turn the forum benchmark into useful evidence.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
