Machining Services vs Alternatives: Which Material-Process Pair Wins Your Part

You do not need to become a process engineer to choose between machining services and a casting, an extrusion, a forging, or a metal 3D print. You need one skill that process brochures never teach: reading the quotes for the same part as one comparable cost story. This guide translates each process into the language of a quote, which cost lines it carries, which line it hides, and how to decide on quantity and geometry before a single bottom line is trusted.

Why is a process choice really a decision about which quote you can read?

Most process comparisons start from the brochure. Casting pages talk about complex near-net shapes, forging pages about grain flow, extrusion pages about constant cross-sections, and metal additive pages about design freedom with no tooling. All of that is true, and none of it tells you what the part will cost at your order quantity. A brochure describes what a process can do. A quote describes what owning that process for one part profile will cost, and ownership, not capability, is what you decide.

The clean way to see this is to send the same drawing to five suppliers and ask each for a price at the same quantity: one machining services shop, one die caster, one extruder, one forger, and one metal additive house. Five quotes for one part should be the easiest comparison in sourcing, and it is the most misleading one, because the five documents are not priced the same way. Machining prices machine time, material, and one-time setup. Die casting prices a die, shots, trimming, and the machining that cast parts still need. Extrusion prices a die and a profile run plus cutting and secondary work. Forging prices an expensive die and a near-net blank that still visits a machine tool. Additive prices expensive machine hours with no die at all. Each quote is a point on a different cost curve, and a bottom-line comparison across them compares points that were never on the same graph.

A "material-process pair" is simply the unit you actually buy: the process that makes the preform plus the process that finishes it — a die-cast housing with machined bores, an extruded profile with machined mounting faces, a forged blank with machined datums. When you compare machining services against alternatives, you are comparing pairs, and every pair has its own cost-line structure. Learn the structure of each and the decision stops being engineering judgment. It becomes arithmetic you can check against any quote.

What does each process quote list, and where is the trap line?

Five processes, five different bundles of cost lines. The table below shows what a typical quote for each process actually lists, and the line buyers most often overlook when they compare bottom lines.

Process	Cost lines a typical quote lists	The line buyers most often miss
CNC machining services	Machine time × rate, material, one-time programming and fixturing, finishing, inspection	That the one-time programming and fixturing line dominates at low quantity — the same part falls sharply in price as the batch grows, so a per-piece comparison against a high-volume process is not an apples-to-apples read.
Die casting	Die tooling amortized over the run, per-shot cycle cost, material, trimming, secondary machining of datum faces	Trial-shot scrap, die maintenance, and the die's limited life — plus the reality that every tight tolerance, tapped hole, and sealing face on the drawing is a separate machining line that a casting quote may not have priced at all.
Extrusion	Low-cost die, per-meter profile price, cutting to length, straightening	Straightness and twist correction on long thin profiles, the minimum economical press run behind a low die price, and the machining every precision face, hole, or slot needs after the profile is cut.
Forging	Expensive dies, forging operation, heat treatment, machining of the blank	The minimum batch economy that makes a forging die viable — below it, the die cost per part is brutal — and the fact that grain-flow quality is bought but only proven by process and inspection records, not by looks.
Metal additive manufacturing	Build time at a high machine rate, powder, build setup, support removal, heat treatment, final machining of datums	That post-processing routinely adds 30–40% to the part's cost; a per-gram price compares powder only, while support removal, stress relief, and datum machining are where the real bill lands.

Notice the pattern running through the table: the machining line survives every alternative. Castings are trimmed and then machined on their critical faces, extrusions are cut and then machined where anything precise must fit, forgings are machined to final dimensions, and metal prints are machined on their datums. Machining services are not only a process among five. They are also the finishing language that four other processes must speak at the end. That is why the comparison that matters is rarely "machine the whole part or use another process for the whole part." It is "how much preform cost does the alternative save, and how much machining does it still owe me?"

The second pattern is the trap line in the right-hand column. A casting quote that does not name trial shots, trimming, and post-machining is a quote for the happy parts only. An additive quote that lists powder and build hours but not support removal and stress relief is a quote for half the part. The quote-reading discipline from this site's machining services cost model — convert every total back into its named cost lines before you trust it — applies across processes, not just across machining suppliers.

Which two questions reduce every part to a shortlist?

Once you can read the cost lines, most process choices collapse into two questions you can answer from the drawing and a demand forecast, before any supplier is called.

Question one, which side of the tooling crossover does your quantity sit on? Every process with a die or a mold carries a one-time cost that must be divided by the parts that use it. Machining has essentially no die, so its per-part cost starts low and falls gently. Die casting, forging, and injection molding have expensive tooling, so their per-part cost starts astronomically high at one piece and falls hard as quantity climbs. The quantity where the two curves cross is the crossover. Below it, no-tooling processes — machining, and for suitable geometry extrusion — win on total cost. Above it, the tooled process takes over, and a widely used benchmark puts the die-casting crossover somewhere near ten thousand units, with the exact number set by die price, cycle time, and how much machining the cast part still needs.

Question two — is the geometry inside the process's constraint box? Quantity is irrelevant if the shape cannot be made. Casting needs draft on walls, a practical minimum wall thickness, and no long thin features that the metal cannot fill. Extrusion needs a cross-section that is constant along the full length and open to being pulled straight through a die. Forging needs shapes a two-part die can form, and economics that tolerate a big die for small runs. Additive accepts almost any geometry but pays for it in machine hours and post-processing. If the geometry violates the box, the process is disqualified regardless of quantity, and the shortlist shrinks without any pricing discussion.

The two questions combine into four practical outcomes, and each outcome points at a different set of quotes to collect:

• Geometry fits, quantity below the crossover. Machining wins on total cost; extrusion wins if the part is a long constant-section profile that machining would have to hog out of solid. Collect machining quotes, and an extrusion quote if the profile rule fits.
• Geometry is boxed in, quantity below the crossover. The design is fighting its own process. Either the geometry changes — draft added, walls thickened — or machining and additive carry the part until volume justifies a tooled process. Collect machining and additive quotes so the price of the constraint is visible.
• Geometry fits, quantity above the crossover. Die casting or forging takes the bulk shape, and machining takes the features the preform cannot hold. Collect the tooled-process quote and the post-machining quote together, because the pair, not the preform, is the unit of comparison.
• Geometry is boxed in, quantity above the crossover. This is the hybrid case: the tooled process makes the body that fits, and every feature outside its box is machined afterward. The part profile splits into as-processed and as-machined requirements, and the winning quote is the one that prices both halves honestly.

Three worked examples that show where the crossover sits

All figures below are example numbers chosen to show the shape of the arithmetic, not prices to budget against. Real tooling costs move with part size, die complexity, and the market that builds the die, and per-piece costs move with the drawing. Read the examples for the structure they teach, then get real numbers for your part.

Example 1 — 500 aluminum housings a year. A machined housing from billet carries no die cost at all; the price is machine time, material, and setup spread over 500 pieces, with the programming and fixturing one-time cost spread thin by comparison. Now price the die-cast alternative: with tooling in the $25,000–$40,000 range, the die alone costs $50–$80 per piece before the first casting is poured. Add shot cost, trimming, and the machining of the datum faces and sealing bores, and the cast route is not competitive at 500 pieces — the die amortization is a wall that quantity never climbs. The crossover sits somewhere far above this volume for a part of this size. That is the honest answer to "when does casting beat CNC": not at this quantity, and the arithmetic shows why without a single engineering judgment.

Example 2 — the same housing at 50,000 pieces a year. Spread the same $25,000–$40,000 die over 50,000 parts and the tooling line drops to about $0.50–$0.80 per piece — inside a dollar, a rounding error next to the machined part's machine time. The die-cast body now wins the bulk cost decisively. The machining that was once the whole part shrinks to the features the casting cannot hold — the bore that must sit at ±0.003 mm (milling) / ±0.002 mm (turning), the tapped holes, the flat sealing faces — each still machined at the tolerance the drawing demands. What keeps CNC alive here is not the body but the machining content. A housing whose drawing is mostly tight features stays mostly a machining job even at 50,000 pieces, because casting holds ±0.1 mm at best and every critical dimension is a machining line. A housing with a handful of machined features becomes a casting with a machining subcontract. The pair, again, is the unit that decides.

Example 3 — thin heat-sink fins the process boxes cannot form. Take a heat sink whose fins are thin, tall, and change direction across the base, with flatness required on the mounting face that the fins sit against. Extrusion is disqualified first: the cross-section is not constant along the length, and a profile die pulls one shape, not a changing one. Die casting fights the fins as well: it asks for draft on every wall and a practical minimum wall thickness, and thin tall fins trap heat in the die and resist filling. The flatness spec on the base is also far tighter than the ±0.1 mm a die-cast surface can promise. The shortlist collapses to machining and metal additive. Machining cuts the fins from solid and holds the flatness; additive forms internal channels or lattice that no cutter can reach, then pays for support removal, stress relief, and datum machining. Volume now decides inside the shortlist: low volumes and thermal-critical faces point to CNC; a design whose value lives in internal channels no tool can reach points to additive. Both quotes should sit on your desk, because the geometry already eliminated the other three processes.

Why does the crossover move every time your design changes?

The crossover quantity is not a fixed property of a part. It moves with the number of design revisions the part will suffer over the tooling's life. Every revision that changes the envelope can demand a die modification or a new die, and a modified die is a new one-time cost that must be amortized all over again. Compare two buyers ordering the same 30,000-piece housing. One freezes the design and orders for three years: the die amortizes once and the crossover is crossed in year one. The other improves the housing every twelve months, and each revision touches the envelope. The die is re-cut or reworked annually, the tooling line repays every year, and the effective crossover never gets reached — the cumulative cost curve for the no-tooling machining route and the tooled route never converge, because the tooled route keeps paying its entry fee.

This is the five-year view that a single RFQ cannot show you, and it is why the design-revision forecast belongs in the process decision alongside quantity and geometry. A part that will often be revised, or that exists in many variants needing their own tooling, is a part where machining's no-die structure is a genuine asset: the program changes, the fixture changes, the per-part cost barely moves. A part whose design is frozen and whose demand is steady is the part that should cross the line into a die, because the die is then an asset doing its job. The mistake is deciding the process on a one-year quantity while the product runs for five. The honest RFQ asks the supplier for the volume forecast horizon, not just the first order, because the crossover answer changes with it.

How do you write a CNC-to-volume handoff into the RFQ?

The smart buying sequence for most parts is not "pick one process and commit." It is: order the first batches in CNC to validate the design, then move the frozen part to the volume process once the crossover is genuinely crossed. The mistake is treating that as two unrelated purchases. Write both legs into the same RFQ, and ask every candidate to price them as one program.

Leg one is the machining services order — the quantity you need now, priced with the one-time programming and fixturing named as a line, and the tolerance and finish callouts written so the shop can price them. Leg two is the transfer line. On the same drawing, ask the supplier to state what would change if the part moved to a cast or forged blank at volume, which walls need draft, which thicknesses must grow, where the parting line would sit, and which features would still be machined and at what tolerance. Then ask what the tooling would cost, how many shots it would live for, and which qualification steps — first article, sample approval, material and porosity checks — gate the transfer. A machining shop that does not cast cannot quote the die, but it can quote the machining half of the pair and tell you honestly what a cast blank would need. That written answer is your transfer line, and it belongs in the quote comparison as much as the unit price does.

Running a mixed program — a CNC first order now and die-cast repeat orders later, sometimes with both lines alive for the same part — demands one quality standard, not two. Split the drawing into what the preform must meet and what the machined part must meet, and inspect both: incoming blanks against their specification, finished parts against the machining callouts, with first-article records for each process and each tool change, and process-capability numbers on the key characteristics. This floor holds ±0.003 mm on CNC milling, ±0.002 mm on CNC turning and reaches ±0.002 mm on precision-ground features at Ppk ≥ 1.67 initial; the blank that feeds those features is verified before it enters the machine, the same way any incoming material is. The quality page on this site lists what that documentation set looks like; buy the records once for the whole program rather than re-deriving them for every process.

One honesty note before you compare: a machining services shop is not a foundry, and a foundry is not a machine shop. Dongguan Zhuohang Ltd, the company behind this site, has machined since 2005 in a 6,800 m² plant with a floor of 100-plus machines and 100+ people at No. 8, Shuibian Songyuan Road, Hengli Town, Dongguan, Guangdong, China 523470, registered as 91441900MA4UWT6X39, with ISO 9001:2015 (116024/A/0001/UK/En), IATF 16949:2016 (131941/A/0001/SM/En), and ISO 14001:2015 (116024/B/0001/UK/En) all in one name. When the winning pair includes a casting or forging, that preform comes from a separate plant with its own certificates and its own process risk — the machining quotation and the foundry quotation are two different documents, and both belong in your comparison, priced and qualified separately.

Write the transition into the RFQ before the first order ships, and you keep both doors open. The machining supplier knows the part will leave at volume and prices the first order accordingly; the foundry knows the machined features are coming and quotes the die honestly; and your audit trail runs from the first machined article to the thousandth cast one without a gap. The alternative — discovering at 20,000 pieces that the die you never priced is the only way to stay competitive, and that the transition needs a redesign, a new first article, and a second supplier — is how process risk becomes a surprise bill. The quote-screening guide on this site shows how to read the supplier's answers to exactly these questions before you commit.

Five questions to answer before the process choice rides on unit price

Let unit price be the last line in the decision, not the first: it is the one figure that assumes everything else is already settled. Answer these five before you let it vote:

• Will quantity really reach the crossover inside the tooling's useful life? A three-year forecast, not a first-order quantity, is the number that should sit under the decision, and the forecast must be honest about how far above the crossover it sits.
• Is the geometry boxed in by the process? Draft, minimum wall, constant cross-section, parting line, die-pull — name the constraint the candidate process would impose, and check the drawing against it before collecting quotes, not after.
• Which tolerances and finishes would each process outsource? Every process hands some features back to machining or to a finisher. List the features that must be machined or finished after each preform, and price those lines with the preform, not beside it.
• How often will this design change? Each revision that touches the envelope re-opens the tooling cost. A revision cadence faster than the amortization horizon is a strong argument for a no-die process, whatever the volume forecast says.
• Who owns the process risk, and is that owner named in the comparison? Someone must translate the geometry into a process, split the drawing into preform and machined requirements, and answer for the transfer. If the supplier's quote does not name its assumptions, the risk has no owner; a comparison that cannot say who decided the process cannot be audited.

What this shop actually runs

If you ask for capability evidence, you deserve a specific answer. This facility operates 100+ CNC machines across 6,800 m2 in Dongguan, China, staffed by 100+ people. The machines, listed by category:

• 5-axis machining centres (10+) - Mazak VARIAXIS I-600 (510/910/510 mm travel, 12,000 rpm, 22 kW, 71.6 N·m, 30-tool BT40) and Mazak INTEGREX I-250H S (B-axis 0.0001°, 24 kW, 124 N·m, 38-tool HSK-A63). Used on compound-angle parts, single-setup multi-face prismatic work, and aerospace fittings where datum transfer loss is unacceptable.
• 3-axis / 4-axis machining centres (50+) - Brother SPEEDIO S500Z2N / S700Z2N (10,000 rpm, 14/21-tool ATC, 0.7 s tool-to-tool, Z-axis 2.2 G) and Mazak VCE-430A L / VCE-570C L (10,000 rpm, 118 N·m, 24-tool ATC). Used on prismatic parts, plates, brackets, housings.
• Swiss-type turning centres (10+) - Citizen Cincom A20-3F7 (Φ20 mm, 10,000 rpm), TSUGAMI B0206-III / B0205-III (Φ20 mm, 0.1 µm), STAR SR-32JII / SR-32J Type N (Φ32 mm, 8,000 rpm). Used on long, slender rotational parts at ±0.002 mm.
• CNC turning centres (20+) - HEADMAN T70M/1250 (720 mm swing, 117 mm bore, 3,000 rpm, BMT75 12-station turret). Used on shafts, bushings, flanges up to ±0.002 mm.
• Wire EDM (5+) - Sodick AQ535L (550/350/300 mm travel, ±20° taper, 800 kg workpiece). Used on hardened tool steel, internal splines, sharp internal corners that milling cannot reach.
• Inspection: Hexagon CMM + 20+ gauges - Hexagon CMM for coordinate measurement, plus vision system, profilometer, hardness tester, coating thickness gauge, XRF for material verification. The 10:1 rule (gauge 10x more accurate than the band) is applied to every feature.
• Machine highlights - VARIAXIS I-600: one-setup multi-face 5-axis + MAZA-CHECK auto-calibration; INTEGREX I-250H S: DONE-IN-ONE mill-turn, B-axis 0.0001° + Ai thermal shield; SPEEDIO S500Z2N / S700Z2N: Z-axis 2.2 G + 0.7 s ATC; Swiss-type: guide-bush switchable + 0.1 µm resolution.

Tolerance bands: ±0.003 mm milled, ±0.002 mm turned, ±0.002 mm ground or wire-cut. Capability indices: Ppk >= 1.67 at the initial study; Cpk >= 1.33 ongoing in production.

Certifications: ISO 9001:2015 (116024/A/0001/UK/En) · IATF 16949:2016 (131941/A/0001/SM/En, design clause 8.3 excluded) · ISO 14001:2015 (116024/B/0001/UK/En), held by Dongguan Zhuohang Ltd (USCC 91441900MA4UWT6X39, since 2005).

Service area: we supply Austria, Belgium, the Czech Republic, Denmark, France, Germany, Greece, Hungary, Ireland, Italy, the Netherlands, Spain, Sweden and the United Kingdom in Europe; Brazil, Canada, Mexico and the United States in the Americas; China, Israel, Japan, Singapore and Sri Lanka in Asia; Australia and New Zealand in Oceania; and South Africa in Africa, with worldwide delivery from China.

Contact: Felipe Iglesias, [email protected], +86 13126536955.

Frequently Asked Questions

Standards referenced on this page

• ASTM A484/A484M-24 — Standard Specification for General Requirements for Stainless Steel Bars, Billets, and Forgings
• ASTM B211/B211M-19 — Standard Specification for Aluminum-Alloy Rolled or Cold Finished Bar, Rod, and Wire
• ISO 9001 — Quality management systems — Requirements